Journal of Biogeography (J. Biogeogr.) (2007) 
w 
ORIGINAL 
ARTICLE 
Effects of rain forest logging on species 
richness and assemblage composition of 
small mammals in Southeast Asia 
Konstans Wells1*, Elisabeth K. V. Kalko1'2, Maklarin B. Lakim3 and Martin 
Pfeiffer1 
'Department of Experimental Ecology, 
University ofUlm, Albert-Einstein Allee 11, 
D-89069 Vim, Germany, 2Smithsonian 
Tropical Research Institute, Balboa, Panama 
<W ^&%W Park, Pen Swraf J0626, 88806 
Kota Kinabalu, Sabah, Malaysia 
"^Correspondence: Konstans Wells, Department 
of Experimental Ecology, University of Ulm, 
Albert-Einstein-Allee 11, D-89069 Ulm, 
Germany. 
E-mail: konstans.wells@uni-ulm.de 
ABSTRACT 
Aim The effects of logging and habitat degradation on the richness and 
abundance of small mammals in Asian rain forests are largely unknown. This 
work compares the species richness, dominance and evenness of small non-volant 
mammals between logged and unlogged forests, and assesses whether assemblage 
variability (^-diversity) is similar between forest types. 
Location Southeast Asia, northern Borneo (Sabah, Malaysia), Sunda-shelf. 
Methods We surveyed species-rich assemblages of small non-volant mammals 
in three unlogged and three logged forests for 2 years. At each forest site, we 
sampled a permanently marked transect and two additional sites in three trapping 
sessions. All analyses were performed at both levels to include the effects of local 
abundances and point estimates, separately from the relative abundances of 
species on a more regional scale. 
Results We trapped a total of 1218 individuals of 28 species. Eleven common 
species accounted for 95% of all captures. Species richness and diversity were 
significantly higher in unlogged forest (27 species) than in logged forest (17 
species). This was mainly attributable to the smaller number of rarely recorded 
species in logged forest (five compared with 16 in unlogged forest, with a total of 
fewer than 10 captures). However, all common species were present in both 
logged and unlogged forests, and our analyses revealed similar patterns of 
dominance, evenness and fluctuations in abundance. Hence overall assemblage 
composition in multivariate space did not differ greatly between forest types. 
Assemblages of Muridae and Tupaiidae showed similar population fluctuations in 
space and time, indicating that the ecology of these taxa may be partially driven 
by the same environmental factors. 
Main conclusions Although species were distributed patchily within sites, 
analyses at local and regional scales revealed similar patterns in diversity and 
assemblage variability, suggesting that effects of forest modification did not differ 
extensively locally and regionally, but had a profound effect on rare species. Our 
results emphasize the importance and conservation value of logged forest stands 
that are able to hold a large proportion of the small mammals also found in 
unlogged forests. Rare and more specialized species are more vulnerable to forest 
degradation than commonly caught species, resulting in the complete loss, or a 
decrease in numbers, of certain groups, such as arboreal small mammals and 
Viverridae. 
Keywords 
Community structure, forest structure, logging impacts, Sabah, small mammals, 
spatial scale, Sunda region, Tupaia. 
? 2007 The Authors 
Journal compilation ? 2007 Blackwell Publishing Ltd 
www.blackwellpublishing.com/jbi 
doi:10.1111/j. 1365-2699.2006.01677.x 
K. Wells et al. 
INTRODUCTION 
The species diversity and structure of local assemblages of rain- 
forest animals are influenced by many factors, including 
habitat complexity and patch heterogeneity, leading to differ- 
ences in the spatio-temporal availability of resources. Tropical 
rain forests, in which tree diversity provides the essential 
resources of food and structural heterogeneity, consist of 
dynamic patches that are frequently affected by local distur- 
bances such as tree-fall (Denslow, 1995; Condit et al, 2000; 
Schnitzer & Carson, 2001). The impact of naturally occurring, 
low-to-intermediate disturbance levels is not necessarily 
negative; it can also enhance diversity through an increase in 
heterogeneity and patchiness of the environment (intermediate 
disturbance hypothesis: Connell, 1978; Molino & Sabatier, 
2001). However, large-scale and high-intensity disturbances, 
which are prominent in many commercially logged forests, 
often have negative overall effects on species assemblages, even 
though logging can resemble naturally occurring large tree-fall 
gaps if reduced-impact logging is applied (Sist et al, 2003). 
Given the ever-increasing anthropogenic pressure on natural 
environments, and in view of the continuing degradation of rain 
forests, ecologists and conservationists face a growing challenge 
in trying to understand more fully the effects of human land-use 
on species, assemblages and ecosystem functioning. Borneo 
contributes considerably to the high biodiversity of Southeast 
Asia (Myers et al, 2000), although many functional groups of 
flora and fauna have only rarely been subjected to detailed 
studies (Sodhi & Liow, 2000). Deforestation in this area is 
progressing more rapidly than in any other rain-forest biome 
worldwide (Curran et al, 2004; Sodhi et al, 2004). Most forests 
in Borneo will probably be logged in the foreseeable future, 
leaving the largest proportion of land area either deforested, or 
covered with logged forest of reduced economic and ecological 
value. The recovery and succession of the remaining forest 
stands depend on the availability of plant resources such as seeds, 
and on the presence of herbivores and their predators (Howlett 
& Davidson, 2003; Brearley et al, 2004). Following the loss or 
alteration in numbers of resources, species or functional groups, 
the interactions of the remaining species will probably be 
negatively affected. 
Because of their seed and seedling consumption, small 
mammals are assumed to play a central role in changes 
occurring within logged rain forest in terms of the composi- 
tion and succession of plant species (Asquith et al, 1997; Blate 
et al, 1998; Lambert et al, 2005; Wells & Bagchi, 2005). 
Increases in the abundance of some small mammals can lead to 
an increase in seed predation, which may suppress forest 
regeneration (Struhsaker, 1997). Moreover, the consumption 
of plant material and herbivorous arthropods by small 
mammals will have a further impact on plant regeneration. 
As small mammals are versatile and exploit tropical rain forests 
in three-dimensional space, including a wide variety of habitat 
patches that are inaccessible to larger vertebrates (Bourliere, 
1989), they may even influence forest structure in areas that are 
at an early stage of succession. 
Although the effects of forest fragmentation on small non- 
volant mammal assemblages (Laurance, 1994; Lynam & Billick, 
1999; Goodman & Rakotondravony, 2000) and the surround- 
ing matrices (Gascon et al, 1999; Pardini, 2004) have been 
intensively studied, less work has been done that addresses the 
effects of logging on small-mammal communities in Asian 
tropical forests (Wu et al, 1996; Laidlaw, 2000; Yasuda et al, 
2003; Bernard, 2004) or elsewhere in the tropics (Laurance & 
Laurance, 1996; Malcolm & Ray, 2000; Lambert et al, 2005). 
As a result, our current knowledge of the impact of habitat 
degradation on small-mammal assemblages is still poor for 
many areas, particularly for dipterocarp rain forests. Most 
studies in the dipterocarp rain forests of Southeast Asia have 
included only two sites (logged vs. unlogged) and no 
replication, although it is becoming increasingly evident that 
the effects of habitat degradation on species diversity and 
animal dynamics are strongly dependent on the spatial scale of 
sampling and landscape heterogeneity (Condit et al, 2002; Hill 
& Hamer, 2004). These aspects thus require multi-site 
approaches and a larger sampling effort. 
Within unlogged dipterocarp rain forests, a large proportion 
of small mammals are generalists (e.g. some abundant murids 
and tupaiids) with overlapping dietary composition and 
microhabitat use (Langham, 1983; Emmons, 2000; Wells et al, 
2004, 2006). This large overlap and flexibility in habitat use 
might blur possible scale-dependent effects on the occurrence 
of species within a heterogeneous forest matrix, although 
patchiness in the distribution of key resources might influence 
demography, even in generalist species (Adler, 2000). The 
demography of small mammals is also expected to differ along 
habitat gradients according to the degrees of specialization and 
colonization ability of the species (Seamon & Adler, 1996), 
including preferences for small-scale perturbations such as 
tree-fall gaps (Beck et al, 2004). 
The floral composition of logged forests differs from that of 
unlogged forests. These differences affect not only overall 
structure and resource availability, but also the scale of habitat 
heterogeneity (Cannon et al, 1998). On one hand, logging 
may lead to simultaneous changes in species composition and 
a strong increase in species that are tolerant to logging-induced 
habitat changes (Cottingham et al, 2001); on the other hand, 
species composition might change but compensatory changes 
in species populations might maintain certain community 
properties, such as overall abundance, at a relatively constant 
level (Ernest & Brown, 2001). Whether the effects of habitat 
alterations are largely compensated for at the local or regional 
level, or whether they are chiefly influenced by habitat 
conditions, depends on the extent of environmental perturba- 
tion and tolerance of key species (Brown et al, 2001). The 
assessment of what species persist in logged forests, and an 
examination of whether and under what conditions certain 
properties of small-mammal assemblages are maintained (and 
at what level) are therefore particularly important. The 
collection of these data may prove crucial in unravelling 
whether, and in what manner, functional groups or taxa 
respond to logging, eventually allowing generalizations about 
2007 The Authors. Journal compilation 
Journal of Biogeography 
2007 Blackwell Publishing Ltd 
Logging impact on Bornean small mammals 
particular sets of species and the creation of conservation 
strategies that preserve as many species as possible. 
To fill these gaps in our knowledge of the species richness 
and assemblage dynamics of small mammals in Southeast Asia, 
we compared species richness, dominance and evenness of 
small non-volant mammals between logged and unlogged 
forests, and assessed whether assemblage variability (P-diver- 
sity) is similar between forest types. Furthermore, we inves- 
tigated factors that lead to changes in assemblage composition 
and analysed the ways that specific taxonomic groups 
(Tupaiidae and Muridae) react to forest degradation. 
METHODS 
Study area 
Borneo is the second largest tropical island after New Guinea. 
It harbours a diverse flora and fauna of approximately 3000 
tree species (MacKinnon et al., 1996) and around 130 non- 
volant mammal species (Payne et al., 1998), comprising both 
Sundaic elements and a high degree of endemism. The moist 
tropical climate, with annual average rainfall of 2670 mm and 
annual mean temperature of 26.7?C (Danum Valley Conser- 
vation Area), is characterized by two periods of pronounced 
rainfall in May-June and October-January (Walsh & Newbery, 
1999). Droughts induced by El Nino events affect fruit 
production by triggering synchronous fruiting of dipterocarp 
trees, a key resource for many vertebrates (Curran & Webb, 
2000). Despite extensive and ongoing clearance, the remaining 
forest areas at present are estimated to cover half (48%) of the 
land surface; this includes large forest patches (Sabah Forest 
Department, personal communication). However, extensive 
areas of forest have previously been logged at least once and are 
thus altered to some extent. 
We selected a total of six study sites: three in unlogged 
lowland rain forest and three in logged lowland rain forest, all 
situated in northern Borneo (Sabah, Malaysia) at altitudes of 
200-900 m (Fig. 1). All forest stands were at least 1000 ha and 
between 17 and 236 km (mean 130 ? 80) apart. Unlogged 
forests: Danum Valley Conservation Area at 04?57' N, 
117?48'E, 'Ufl'; Poring, Kinabalu NP at 06?02'N, 
116?42'E, 'Uf2'; Tawau Hills NP at 04?23'N, 117?53'E, 
'Uf3'; logged forests: Luasong Field Centre at 4?36' N, 
117?23' E, Til'; Kg. Monggis at 06?13' N, 116?45' E, 'Lf2'; 
Kg. Tumbalang at 06?08'N, 116?53' E, 'LB'. Whereas the 
unlogged forest stands were characterized by relatively undis- 
turbed vegetation, with emergent trees rising up to 60 m, 
canopy heights at the logged forest sites, which had been 
selectively logged c. 15-25 years prior to our study, reached 
only c. 25-30 m, and the few remaining larger trees (e.g. Ficus 
spp.) were of no commercial value. Harvesting practice in the 
highly productive Bornean forests may exceed 10 trees ha-1, 
damaging more than 50% of the original stand (Sist et al, 
2003). 
Details of the logging histories at the various sites were not 
available, although all logged sites were structurally similar. 
Lfl 
1C: km 
Uf3 
Figure 1 Map of Borneo with the six study sites. Unlogged for- 
ests: Danum Valley Conservation Area 'Ufl'; Kinabalu NP 'Uf2'; 
Tawau Hills NP 'Uf3'; logged forests: Luasong Field Centre 'Lfl'; 
Kg. Monggis 'Lf2'; Kg. Tumbalang 'Lf3'. 
Half the original stands were damaged to a certain degree, as is 
usual in conventional logging practice where trees are cut and 
then transported through the unlogged areas to nearby roads. 
Compared with unlogged forest, disrupted canopies and 
pronounced gaps in logged forest lead to differences in plant 
composition and understorey, in which fast-growing plants 
such as climbing bamboo (Dinochloa spp.), sago palms 
(Metroxylon spp.) or rattan (Calamus spp.) dominate. The 
hunting of most vertebrates is common in nearly all forest 
areas. Small, non-volant mammals in particular are hunted in 
logged forests because of the large number of rural villages 
nearby and easy access via logging roads (K.W., personal 
observation). However, these influences were assumed to have 
little effect on our sampling, as the hunting of small mammals 
did not take place near our study locations. 
Animal capture and handling 
Animals were captured with locally made wire-mesh live traps 
equipped with a plastic roof for rain protection 
(280 x 140 x 140 mm) between September 2002 and Novem- 
ber 2004. We conducted 18 trapping sessions, alternating 
between six study sites with a mean interval of 103 ? 50 days 
between consecutive trapping sessions at the same forest site, 
giving a total of three sampling units per site. Trapping was 
carried out at equal intervals throughout the seasons and year 
for both forest types (Ufl: Mar 03, Dec 03, Sep 04; Uf2: Dec 02, 
Jun 03, Feb 04; Uf3: Oct 02, Aug 03, Apr 04; Lfl: Apr 03, Sep 
03, May 04; Lf2: Nov 02, May 03; Jan 04; Lf3: Jul 03, Mar 04, 
Nov 04). At each forest site, we established a randomly located 
and permanently marked transect of 40 traps set 20 m apart in 
two parallel lines on the forest floor. Additionally, 60-116 traps 
Journal of Biogeography 
? 2007 The Authors. Journal compilation 2007 Blackwell Publishing Ltd 
K. Wells et al. 
were placed arbitrarily at two additional locations about 0.5- 
1.1 km away from the transects, except in Poring where the 
distance was only 300 m because of the topography. Addi- 
tional traps were set at various places to enhance data on 
species diversity. The traps were baited with banana and 
checked every morning for 16 consecutive days during each 
trapping session. 
The mean trapping effort was 2148 ? 408 trap-nights (traps 
active for 24 h) per session. Captured animals were briefly 
anaesthetized (we used diethylether, which had no apparent 
long-term impact on animal sedation), marked with a pit tag 
(ARE 162 transponders, AEG, Germany), measured, then 
released at the point of capture. Although some rats at 
arbitrarily sampled sites were marked only by ear punching, 
they could be identified reliably as recaptures during the entire 
study period. Species identification was based on Payne et al. 
(1998) and on comparison with specimens from the Sabah 
Parks Museum (Kundasang, Sabah, Malaysia). Specimens of 
accidentally killed animals, or individuals collected during the 
last trapping sessions, were deposited either at the Sabah Parks 
Museum or at the Senckenberg Museum, Frankfurt, Germany. 
Data analysis 
We examined species richness in logged and unlogged forests 
by using sample-based rarefaction curves based on the total 
trapping effort throughout the study period. This approach 
incorporated spatial heterogeneity, which is inherent to almost 
all samples (Colwell et al., 2004). We then estimated the 
expected number of species for logged and unlogged forest 
with Chao2 and lackl species-richness estimators (Colwell 
et al, 2004). 
We established data matrices that included the number of 
individuals of each species trapped in transects during the 
16-day periods (483 individuals in 18 transects, 'trans') and the 
first 54 individuals from a trapping session pooled from 
captures in transects and additional locations ('sess'). Two 
trapping sessions were excluded from analysis because of small 
sample size, leaving 864 individuals in 16 sessions for analysis. 
We analysed our data at the assemblage level and with respect 
to two taxonomic groups, murids and tupaiids, which 
dominated in all of our samples. All analyses were performed 
at both levels to include the effects of local abundances and 
point estimates (trans), separately from the relative abun- 
dances of species on a more regional scale (sess). As some 
immature spiny rats (Maxomys rajah and M. surifer) were not 
distinguishable in the field, we added the unidentified 
individuals proportionally to the number of identified indi- 
viduals of both species for analyses (45 out of a total of 171 
individuals, Table 1). 
For each matrix, we used the coefficient of variation 
(CV = SD/average number of individuals) to describe varia- 
tions in species abundances within sites, and the non- 
parametric Shannon-Wiener index H' to analyse species 
diversity (Magurran, 2004). Bray-Curtis (quantitative Soren- 
sen)  similarity matrices were used for comparisons across 
species assemblages from local and regional estimates. Based 
on these matrices, we conducted a two-dimensional non-linear 
ordination with multidimensional scaling (NMDS), which is a 
particularly robust ordination technique (Clarke, 1993). Axes 
scores for axis 1 of the NMDS were correlated to features of 
assemblages and taxonomic groups, respectively, using Spear- 
man's correlation. We tested similarity matrices of species 
assemblages for possible relationships of temporal (chronolo- 
gical time differences in days), seasonal (shortest time differ- 
ences between months of respective sessions), and geographical 
(kilometres between sites) distances between all trapping 
sessions by using Mantel statistics with 1000 permutations. 
We approximated species-specific relative persistence rates of 
individuals in consecutive sessions as PR = re x At I 
{N, x Nf_i), where re is the number of recaptured individuals, 
At is the time lag [days] between sessions, Nt is the number of 
individuals in the session, and Nt-i is the number of 
individuals in the previous session. Diversity estimates were 
calculated with the software ESTIMATES (ver. 7.5, R. K. 
Colwell, http://purl.oclc.org/estimates), Mantel tests were 
conducted with PC-ORD 4.0 (B. McCune & M.J. Mefford, 
1999), and further non-parametric statistics [Mann-Whitney 
(MW) U, Kruskal-Wallis (KW) ANOVA, Spearman's correla- 
tion, %2 test] were performed with STATISTICA 6.0 (StatSoft, 
2001). Values are given as means ? 1 SO. 
RESULTS 
a-Diversity in logged and unlogged forests 
During the entire study, we trapped a total of 1218 individuals 
(trapped at 3809 different times in total) from 28 species, 
representing 17 genera from eight families, with a sampling 
effort of 40,552 trap-nights (Table 1). In the unlogged forests, 
we found more species but fewer individuals (27 species, 547 
individuals) than in the logged forests (17 species, 671 
individuals). The only species that was not recorded in 
unlogged forests was the shrew Chimarrogale himalayica, but 
this was captured only once in logged forest. Being an 
insectivore, it was probably not attracted to our bait. 
Accumulation curves indicated that the unlogged forests 
contained a richer small non-volant mammal assemblage, with 
a steeper accumulation curve, than the logged forests (Fig. 2). 
Estimates of predicted species richness were higher for 
unlogged forests, with 29 ? 3 (Chao2) to 32 + 2 (Jackl) 
estimated species, than for logged forests, with 22 ? 6 to 
21 ? 2 estimated species. These estimates confirmed that 
species richness was lower in logged forests (Fig. 2). Estimates 
of species richness for data sets from the transects (Chao2: 
17 + 3 for Uf, 14 ? 1 for Lf; Jackl: 19 ? 1 for Uf, 16 ? 1 for 
Lf) or standardized sessions (Chao2: 27 ? 4 for Uf, 14 ? 0.2 
for Lf, Jackl: 28 ? 3 for Uf, 15 ? 1 for Lf) were in part lower 
than the total number of documented species; this was 
probably the consequence of rare and therefore slowly 
accumulating species that were under-represented in single 
trapping sessions. 
2007 The Authors. Journal compilation 
Journal of Biogeography 
2007 Blackwell Publishing Ltd 
Logging impact on Bornean small mammals 
Table 1 Number of individuals of all species trapped in the various study areas. Total trap effort for each area in parentheses. Additional 
trapping efforts in Uf2 and Uf3 during the period of field work that was not part of the analyses are included to provide a more complete 
species record. 
Latin name Total number 
Unlogged forest (Uf) Logged forest (Lf) 
Species authority of individuals Ufl (7092) Uf2 (7545) Uf3 (8115) Lfl (6469) Lf2 (6040) Lf3 (5291) 
Rodentia 
Muridae 
Chiropodomys major 
Lenothrix canus 
Thomas 1893 
Miller 1903 
1 
2 
1 
2 
- - - - 
Leopoldamys sabanus Thomas 1887 175 9 76 18 12 39 21 
Maxomys baeodon Thomas 1994 5 1 - - 1 1 2 
Maxomys ochraceiventer Thomas 1894 2 - 2 - - - - 
Maxomys rajah Thomas 1894 76 25 1 3 37 9 1 
Maxomys surifer Miller 1900 50 7 4 10 14 9 6 
Maxomys cf. suriferlrajah 
Maxomys whiteheadi Thomas 1894 
45 
73 
25 
12 
4 
10 
4 
26 
3 
18 
8 
3 
1 
4 
Niviventer cremoriventer Miller 1900 265 28 26 9 19 55 128 
Rattus rattus Linnaeus 1758 4 2 - 2 - - - 
Sundamys muelleri Jentink 1879 41 3 1 1 - - 36 
Sciuridae 
Lariscus hosei Thomas 1892 1 - - 1 - - - 
Cattosciurus notatus Boddaert 1785 11 - 5 - 1 4 1 
Cattosciurus prevostii Desmarest 1822 4 - 3 1 - - - 
Sundasciurus brookei Thomas 1892 1 - 1 - - - - 
Sundasciurus hippurus 
Sundasciurus hwii 
Geoffroy 1831 
Thomas 1892 
7 
45 
1 
8 
1 
2 
4 
21 5 5 
1 
4 
Hystricidae 
Trichys fasciculata Shaw 1801 2 2 
Insectivora 
Chimarrogale himalayica 
Erinacaeidae 
Gray 1842 1 
- - - - - 
1 
Echinosorex gymnura Raffles 1822 4 - - 2 2 - - 
Scandentia 
Tupaiidae 
Ptilocercus lowii Gray 1848 1 1 1 
Tupaia gracilis Thomas 1893 24 1 9 4 4 2 4 
Tupaia longipes Thomas 1893 117 28 10 42 23 4 10 
Tupaia minor Giinther 1876 76 1 25 2 - 4 44 
Tupaia tana Raffles 1821 173 22 15 12 15 54 55 
Carnivora 
Viverridae 
Arctogalidia trivirgata 
Paradoxurus hermaphroditus 
Gray 1832 
Pallas 1777 
2 
6 4 
1 1 
2 
- - - 
Viverra tangalunga Gray 1832 3 3 - - - - - 
Total 1218 181 199 167 154 197 320 
Number of species 28 17 19 19 12 12 16 
Eleven species, each represented by more than 20 captures, 
were classified as commonly caught. These species accounted 
for 95% of all captures and were recorded at all forest sites, 
except for Sundamys muelleri. Murids were most abundant in 
both forest types, accounting for 57% and 63% of all captures 
in unlogged and logged forests, respectively. Tupaiids were 
recorded with lower capture rates (31% and 33% of all 
captures) in unlogged and logged forest. 
Shannon-Wiener diversity estimates showed no clear dif- 
ferences between single site estimates from logged vs. unlogged 
forest sites (MW U test, trans: U = 4.0, P = 0.83; sess: 
U = 3.0, P = 0.51) but the overall diversity between both 
forest types was significantly lower in logged forest (trans: 
H'uf = 2.35 ? 0.02, H'u = 2.08 ? 0.05; sess: H'm = 2.43 ? 
0.03, H'u = 2.12 ? 0.04). 
Dominance and abundance of common species 
The most abundantly trapped species recorded in the sessions 
were   Niviventer   cremoriventer   (most   abundant   in   n = 5 
Journal of Biogeography 
? 2007 The Authors. Journal compilation ? 2007 Blackwell Publishing Ltd 
K. Wells et al. 
40 
35 
30 
25 
0    20 
15 
10 
li in 
? 
\i n n n 
200 400 600 
Number of individuals 
800 1000 
Figure 2 Rarefied species-accumulation curves representing the 
average number of species for a given number of captured indi- 
viduals for the entire regional trapping effort (sessions) in un- 
tagged forest (?) and logged forest (O). Triangles and squares 
refer to the estimated number of species based on Chao2 and Jackl 
estimators, respectively. Bars are 95% CI. 
0.45 
0.35 
0.25 
M     0.15 
0.05 
-0.05 
A 
? Uf1 
? ?Uf2 
? ?Uf3 
? Lfi 
D    ? ?Lf2 
a|_f3 
*    *    o 
U A 
* 
o 
t 
? 
? 
* 
? 
j
 8 8 8 a a ft ft   ft 
6       8       10     12 
Abundance rank 
14     16     18 
Figure 3 Mean relative abundances of species pooled over trap- 
ping sessions at various study sites. No significant differences in 
the shape of the rank abundance curves were detectable (all Kol- 
mogorov-Smirnov two-sample tests, P > 0.10). Note that the 
numbers of species are lower than actually recorded in some ses- 
sions because of data standardization. 
sessions), M. rajah (n = 4), Leopoldamys sabanus (n = 3), 
Tupaia longipes (n = 3), Tupaia tana (n = 2) and Maxomys 
whiteheadi (n = 1). The three most abundantly trapped 
species varied within and between study sites. Whereas they 
were mostly the same within a study site (Sorensen index, 
Smean (site) = 0.70 ? 0.19), the abundant species differed 
more often between sites of the same forest type (Smean 
(Uf) = 0.39 ? 0.26, Smean (Lf) = 0.50 ? 0.29) with no obvi- 
ous differences in dominance patterns between Uf and Lf 
(Smew (UfxLf) = 0.42 ?0.27). 
Fluctuations in the abundance of the 11 most commonly 
caught species, as determined by the CV from each site, did not 
differ across sites (MW U test, trans: U < 3.0, P > 0.08; sess: 
U < 3.0, P > 0.08). Mean CVs ranged from 0.65 at Uf3 to 1.14 
at Ufl for transect data, and from 0.33 at Lfi to 1.07 at Lf3 for 
sessions, with no recognizable difference in abundance fluctu- 
ations   between   species   (KW   ANovAtrans,   H10)53 = 12.00, 
0.28; KW ANOVA* Hu 10.73, P = 0.38). Overall, 
the mean abundance fluctuations pooled for the different 
forest types did not differ between logged and unlogged forests 
(MW U = 397.0, P = 0.43). Likewise, the relative abundance 
of the 11 most dominant species did not differ between 
sessions in logged and unlogged forests (MW U = 36, 
P = 0.79), while the overall abundance distributions as 
indicated by rank abundance curves were not distinct between 
logged and unlogged forests (all Kolmogorov-Smirnov two- 
sample tests (n =15) P > 0.10) (Fig. 3). However, mean 
capture frequencies were significantly larger for N. cremori- 
venter and T. tana in logged than in unlogged forest (both MW 
U> 15, P< 0.05). 
Spatio-temporal variation in assemblage structure 
NMDS ordinations of species assemblages and seasonal/ 
temporal   similarities   between   trapping   sessions   extracted 
two-dimensional solutions in which all raw stress factors, cj>, 
were <0.15, indicating that the original relationships in 
matrices were represented sufficiently by the resulting NMDS 
axes (Clarke, 1993). Changes in assemblage structure at 
the local level based on transect data (trans, Fig. 4) were 
significantly correlated with changes at the regional level based 
on data from complete sessions (Mantel test, r = 0.79, 
P < 0.01). 
Although the number of tupaiid species was correlated with 
the composition of assemblages (NMDS axis 1) at the local 
level based on transect data (trans: Spearman -R?=is = -0.60, 
P < 0.009), the number of murid species was correlated with 
species composition at the regional level, based on data from 
complete sessions (sess: Spearman -R?=i8 = 0.68, P < 0.004; 
Fig. 5a). Changes in the regional assemblage composition of 
murids and tupaiids were significantly correlated (Mantel test, 
r = 0.27, P < 0.05), suggesting that part of this pattern was 
1.4 
1.0 
0.6 
? ? Uf1 
A  Uf2 
HT 
Hf Uf3 
? Lfi 
.  A AA o 
o 
? 
O  Lf2 
A   Lf3 
o 
A 
? 
? 
A ? 
? 
A 
8   ?-2 
9-0.2 
-0.6 
-1.0 
-1.4 
-1.5   -1.0   -0.5     0.0      0.5      1.0      1.5      2.0      2.5 
NMDS axis 1 
Figure 4 Multidimensional scaling plot of small non-volant 
mammal assemblages based on local censuses (trans) during the 18 
trapping sessions. 
2007 The Authors. Journal compilation 
Journal of Biogeography 
2007 Blackwell Publishing Ltd 
Logging impact on Bornean small mammals 
(a) 
CO 
Q 
is 0.8 
E3 local assemblages (transects) 
? regional assemblages (sessions) 
(b) 
0.8 
CO 
Q 
?0.4 
0.0 
|::;;:::| local murid assembl. 
HH regional murid assembl. 
I     I local tupaiid assembl. 
? regional tupaiid assembl 
u&i 
T3 
T3 > a> O 
?o a n 
T T3 CO 
CJ) 
2        (5 
Capture success was lowest during fruiting seasons in 
September and October (Fig. 6), although we found no 
significant general impact of seasonal or temporal differences 
on assemblage compositions (Mantel tests, all r < 0.18, 
P > 0.05). Composition of species assemblages as described 
by NMDS axis 1 scores was most similar within sites and less 
similar across sites (site as fixed factor: KW ANOVA axis 1, 
trans: H51S = 14.45, P < 0.02; sess: H5A6 = 14.01, P < 0.02). 
NMDS axis 1 scores of assemblages in unlogged and logged 
forest were indistinguishable from each other at the local level 
(trans: both MW U > 25, P > 0.17). However, NMDS scores 
of assemblage composition at the regional level differed 
between unlogged and logged forest on the first ordination 
axis (sess: MW U = 11, P < 0.03). 
Distribution of species 
The spatial distribution of species was heterogeneous among 
localities (transects and additional locations) within a study 
site. A comparison of observed and expected distributions of 
commonly caught species, based on the total number of 
captures at different locations, revealed significant deviations 
from a random distribution among locations within each site 
in 20 out of 146 cases (/2 > 6.1, P < 0.05). These patterns of 
spatial heterogeneity regarding species distribution were 
similar for unlogged and logged forest, and were most 
pronounced for L. sabanus, Sundasciurus lowii and Tupaia 
gracilis. 
Figure 5 Impact of selected variables on changes in community 
composition in (a) assemblages of all species and (b) murid and 
tupaiid assemblages, on both local and regional scales. Bars rep- 
resent R values from Spearman's correlations for numbers of 
species and individuals and for H' diversity, and r values from 
Mantel statistics for murid/tupaiid assemblages and geographical/ 
temporal distances. Note that we considered only the first non- 
linear ordination with multidimensional scaling (NMDS) axes in 
these figures. ""Significant correlations (P < 0.05). 
driven by similar environmental fluctuations (Fig. 5b). We 
traced this relationship down to the species level and found 
that murid assemblages described by NMDS axis 1 scores were 
correlated with the relative abundances of T. minor and 
T. longipes (sess: both Spearman Rn=l6 > ? 0.75, P < 0.0006). 
Tupaiid assemblage scores (NMDS axis 1) were correlated with 
the relative abundances of N. cremoriventer, M. rajah and 
M. whiteheadi (sess: all R?=i6 > ? 0.50, P < 0.05). Species 
assemblages were also correlated with respective H' diversity 
estimates, but not with number of captured individuals 
(Fig. 5a,b). 
An impact of geographical distance between sites on 
assemblage similarity (quantitative Sorensen distances) was 
detectable for both local and regional species assemblages and 
for murid and tupaiid assemblages (Mantel tests, all r > 0.29, 
P< 0.01). 
Persistence of individuals 
We recaptured  15%  (n = 120 individuals)  of 784 marked 
individuals.  Persistence  rates  of individuals  in  consecutive 
80 
70 
<1> 60 
o b() 
rri 
u 
"o 40 
II) 
_Q ;M) 
h 
z 20 
10 
0 
??s (\ 
\ 
\ 
A\ /} v 
.... s-y 
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 
Figure 6 Number of captures in transects throughout the year. 
Total numbers of captured individuals (?), murids (A) and 
tupaiids (?) are given. Although the trap success for murids and 
tupaiids was not correlated (Spearman R(?=18) = 0.16, P = 0.54), 
and the verification of general seasonal patterns was weak, the 
trapping success for both taxa was lowest during the fruiting 
season in September/October. Note that some months contain 
more than one sample. 
Journal of Biogeography 
? 2007 The Authors. Journal compilation ? 2007 Blackwell Publishing Ltd 
K. Wells et al. 
trapping sessions differed between forest sites (KW ANOVA 
^5,110 = 18.15, P < 0.003), as persistence rates were relatively 
high in Uf3 but low in Lfl. They were marginally lower in 
logged than in unlogged forest (MW U = 1203, P = 0.07). 
The mean persistence rate was highest for M. surifer, M. rajah 
and T. gracilis. However, the means differed only slightly 
between species because of the large variability of persistence 
rates between sites (KW ANOVA H5J110 = 18.15, P = 0.08) 
(Table 2). The longest persistence was recorded for a T. minor 
individual, which had been marked in a previous study in 2001 
(Wells et al, 2004) and was recaptured after 636 days. 
DISCUSSION 
Small mammal diversity in logged and unlogged 
forests 
In tropical forests, a high structural diversity and great 
variability in resources are considered key elements in the 
maintenance of diverse small-mammal assemblages (August, 
1983). Furthermore, occurrence and assemblage patterns of 
small mammals are determined by the degree of specializa- 
tion, flexibility and general demography of the constituent 
species (Adler, 2000). Species flexibility should not only 
ensure persistence and abundance in spatially and temporally 
heterogeneous forest matrices, but also tolerance of logging 
and habitat disturbance. Although rain-forest logging resul- 
ted in a significant loss of rare small mammal species in our 
study, the ubiquitous presence of commonly caught species 
at all forest sites, both logged and unlogged, suggests that 
assemblage dynamics are mainly determined by these species. 
Multiple comparisons of replicates within and between sites 
have revealed that fluctuations in abundance and assemblage 
variability appear to be little affected by logging, raising the 
question of how far synchronous responses to environmental 
fluctuations account for variation in local species assem- 
blages, and whether this can be traced to the same dominant 
species. 
Table 2 Maximum recorded persistence times for commonly 
caught species. 
Max. Site Total number Mean 
persistence of of recaptured persistence 
Species (days) record individuals rate PR ? SD 
L. sabanus 534 US 14 3.8 ? 5.8 
N. cremoriventer 494 Lf3 20 0.5 ? 0.9 
M. rajah 573 Ufl 23 20.8 ? 41.0 
M. surifer 537 US 5 20.1 ? 37.8 
M. whiteheadi 273 Ufl 5 7.4 ? 14.6 
S. muelieri 250 Lf3 1 0.2 ? 0.7 
S. lowii 262 US 3 0.6 ? 1.8 
T. gracilis 287 US 4 19.3 ? 33.1 
T. minor 636 Uf2 6 0.4 ? 1.0 
T. longipes 590 US 9 2.2 ? 2.8 
T. tana 547 US 19 6.9 ? 13.7 
The pronounced decline in species richness and diversity in 
logged forests was mainly attributable to the reduction in rare 
species, whereas commonly caught species of omnivorous 
murids or tupaiids were recorded almost equally often at all 
sites. This pattern is consistent with other studies on small 
non-volant mammals in Australia (Laurance & Laurance, 
1996), Malaysia (Zubaid & Ariffm, 1997; Yasuda et al, 2003) 
and Venezuela (Ochoa, 2000). 
Common vs. rare species 
In our study, species affected by logging could be arranged in 
various functional groups, but no evidence was found for a 
single factor explaining the lack of certain species in logged 
forests. Four of the rare species we recorded only in unlogged 
forests are endemic to Borneo (Chiropodomys major, Maxomys 
ochraceiventer, Lariscus hosei, Sundasciurus brookei), and their 
restricted geographical distribution might be associated with 
less tolerance to environmental variability compared with 
species that inhabit a larger geographical area. Species decline 
in logged forest was most evident in civets, which are known to 
be sensitive to habitat degradation (Heydon & Bulloh, 1996; 
Colon, 2002). The mainly arboreal rats (C. major, Lenothrix 
canus) and squirrels (Callosciurus prevostii, S. brookei) are less 
prevalent in logged forests, which might be because of reduced 
canopy space and altered tree composition and texture (Saiful 
et al, 2001; Yasuda et al, 2003; Wells et al, 2004) compared 
with unlogged forest. The question remains as to whether 
resource availability, or structure and habitat space, is the main 
determinant of reduced species richness in logged forest, and 
whether these proximate parameters affect particular species 
groups more than others. Some studies have suggested that the 
consequences of habitat disturbance differ with the type and 
spatial extent of disturbance. Favourable circumstances, such 
as an increase in herbaceous vegetation, a decrease in canopy 
and sapling density, and more abundant arthropods and fruits 
(Malcolm, 1997; Struhsaker, 1997; Lambert et al, 2003) may 
lead to increases in small non-volant mammal densities in 
disturbed habitats (Malcolm & Ray, 2000; Lambert et al, 
2005). We found an increased abundance of N. cremoriventer 
and T. tana in logged forests. Although it has been suggested 
that T. tana prefers dense undergrowth and gap structures 
(Emmons, 2000; Wells et al, 2004), whether structural features 
or particular resources are important per se remains unclear. 
Furthermore, whether conclusions from gap vs. understorey 
dynamics are applicable to logged forest conditions is also 
uncertain. 
Species richness and resource aggregation 
If the decline in species in logged forests is mainly attributable 
to resource specialization, a consideration of whether the 
occurrence and abundance patterns of species are driven by the 
presence of particular resources and/or by certain patterns of 
resource allocation would be of interest. For instance, tree 
species that play a significant role in overall forest architecture 
2007 The Authors. Journal compilation 
Journal of Biogeography 
2007 Blackwell Publishing Ltd 
Logging impact on Bornean small mammals 
and resource availability are often not randomly distributed 
within tropical forests (Condit et al, 2000). Rather than 
overall forest structure, such patchy distribution patterns of 
key resources have been shown to influence the demography of 
the Neotropical Proechimys rat, which concentrates its activity 
mainly around fig (Ficus) trees, one of its main food resources 
(Adler, 2000). Therefore small mammals with specialized 
feeding habits and a dependence on spatially clumped 
resources seem to be mostly aggregated. Conversely, common 
species with omnivorous diets cope well with a wide range of 
resources and exhibit greater tolerance towards spatio-tem- 
poral resource availability. If the area covered by the spatial 
variability of plant and other resources exceeds the foraging 
areas of generalist feeders, then more specialized species should 
be able to cope more efficiently with a subset of the resources 
in some localities. This, in turn, might lead to a balanced 
overall dynamic of the assemblage, as the number and 
abundance of specialist and generalist species might compen- 
sate each other. High resource diversity and its specific 
distribution in heterogeneous forests therefore should promote 
the presence and abundance of both specialist and generalist 
species. With respect to capture probabilities, the chances of 
capturing a specialized species should be lower because of its 
reduced abundance, and higher for more generalist species. 
Plant composition and distribution differ in logged forests 
(Cannon et al, 1998). The proportion of animal-dispersed and 
mammal-pollinated trees, as well as arthropod assemblages 
that may serve as food sources for small mammals, may be 
affected by logging (Davis et al, 2001; Chazdon et al, 2003; 
Cleary, 2003). For instance, the reduced availability of 
particular fruit resources in logged forests has been reported 
as negatively influencing densities of the fruit-eating mouse 
deer Tragulus spp. (Heydon & Bulloh, 1997). As outlined 
above, such resource alteration should mostly influence the 
occurrence and density of more specialized small non-volant 
mammals. This is in agreement with our results, as analysis of 
commonly caught species reveals some aggregation at the level 
of individuals, but they are also widely dispersed in different 
locations and forest sites. Reduced species richness in logged 
forests probably occurs mostly as a consequence of reduced 
overall species densities and/or lower abundance of rare 
species, as most of the rare species that we trapped only in 
unlogged forests are known also to be present in logged forests 
(L. canus, C. prevostii, Trichys fasciculata, all civets; personal 
observation). 
Tolerance of logging by common species, and the pro- 
nounced prevalence of rare species in unlogged forests, have 
also been found for birds in the same geographical region 
(Sodhi, 2002; Lammertink, 2004). However, other studies 
conducted at different spatial scales with birds and butterflies 
led to contrasting results, with both decreased and increased 
diversity (Hill & Hamer, 2004). 
The amount of intraspecific aggregation of a species within a 
set of assemblages should concomitantly decrease oc-diversity 
and increase ^-diversity (Veech, 2005). Therefore the spatially 
clumped distribution of species should also be considered in 
the interpretation of variability in assemblages both within and 
among forest types. Surprisingly, the variability in assemblages 
from unlogged vs. logged forest, as determined by multivariate 
analysis, differed at neither the local nor the regional level. 
Furthermore, we found no differences in abundances of 
commonly caught murids and tupaiids within a forest type, 
and no evidence for differential impacts of logging on these 
functional groups. However, some fluctuations in the assem- 
blage of both taxa are evident: while the number of tupaiid 
species were associated with assemblage fluctuations at the 
local level, murid species had more influence on assemblage 
fluctuations on a regional scale (Fig. 5a). 
Logging effects on small-mammal assemblages 
Based on the observation that different plant or invertebrate 
taxa respond inconsistently to anthropogenic habitat alteration 
(Lawton et al, 1998; Ricketts et al, 2002), we conclude that 
habitat disturbance in the form of logging may not necessarily 
lead to the synchronous alteration of food availability for 
different groups of small mammals. Unfortunately, the diet of 
murids and tupaiids is not well known, although they are 
thought to include a large range of arthropods and plant 
material (Langham, 1983; Emmons, 2000). However, interest- 
ing differences exist in their morpho-physiological traits 
related to food processing. Tupaiids have weak jaws in 
combination with short intestinal transition times and sim- 
plified colons (Emmons, 1991) that do not allow the process- 
ing of the hard dipterocarp and lithocarp fruits that are 
favoured by murids during the fruiting season in unlogged 
forests (Curran & Webb, 2000; Wells & Bagchi, 2005). 
Surprisingly, although these fruits comprise a key resource in 
unlogged forests, differences in local abundance related to 
habitat disturbance resulted neither in detectable differences in 
murid fluctuations between unlogged and logged forest, nor in 
any asynchronous changes in murid and tupaiid assemblages. 
Nevertheless, some impact of season in relation to fruiting can 
be inferred from the reduced trapping success during the main 
fruiting peak; this time interval also coincides with the main 
reproductive period of murids in unlogged forests (personal 
observation). 
Another factor contributing to the observed assemblage 
structure patterns in our study could be the geographical 
locality and the distance between study sites. Geographically 
distinct areas differ in climate, altitude and edaphic factors that 
influence plant and resource composition on a regional scale 
(Ashton & Hall, 1992; Newbery et al, 1996). Three of the sites, 
one unlogged and two logged (minimum distance between 
sites 17-24 km; Uf2, Lf2, Lf3), were close to Mount Kinabalu, 
a mountain that strongly influences the topography, soil 
mineral content and climate of this region (Kitayama, 1992). 
Such geographically related factors might be of greater 
importance than factors sensitive to logging in influencing 
abundance fluctuations and assemblage dynamics. Overall, the 
similarities in assemblage features between forest types suggest 
that fundamental ecological or abiotic features of the biome, 
Journal of Biogeography 
? 2007 The Authors. Journal compilation 2007 Blackwell Publishing Ltd 
K. Wells et al. 
rather than profound differences between unlogged and logged 
forest, are major driving forces in shaping assemblage structure 
and abundance patterns. 
Although logged forests are generally characterized by 
distinctly altered plant composition and physical structure 
compared with unlogged forest, many kinds of logging damage 
might in some ways be equivalent to the naturally occurring 
perturbations and alterations to which a large proportion of 
common non-volant small mammal species are well adapted. 
Most of the commonly caught species from our study, such as 
L. sabanus, M. surifer or M. whitheadi, have inhabited a wide 
geographical range in the Sunda region of Southeast Asia 
throughout their evolutionary history (Gorog et al, 2004). 
This supports the idea of the long-term adaptation to, and 
tolerance of changes in, habitat conditions by these species. 
CONCLUSIONS 
Logging does not appear consistently to cause strong changes 
in species assemblages with respect to ubiquitously present 
generalist species. We know little about the multiple interac- 
tions of small non-volant mammals with other components of 
the ecosystem. Further work is required to determine whether 
fundamental ecosystem processes in logged forests are altered 
by changes in resource availability, competitors or carnivorous 
predators, even if the same small mammal species are present 
(Terborgh et al, 2001). The role of rare species remains even 
more elusive, although the reduced species richness in our 
study clearly suggests that some species are vulnerable to severe 
population reductions or extinction by logging-induced chan- 
ges. The inconsistency in logging responses among species, and 
the large habitat variability that is intrinsic to rain forests and 
that is further generated by various anthropogenic impacts, 
present a challenge when selecting areas for conservation. 
Hitherto, general statements on logging effects can be made for 
different species groups. Although logged rain forests might 
house large proportions of the small-mammal assemblages 
found in undisturbed forests, some rare species will remain 
unprotected if unlogged forests are not conserved, as these 
forests remain the major source of the region's immense 
species richness. 
ACKNOWLEDGEMENTS 
We thank the Economic Planning Unit, Malaysia, for a 
research permit, and Sabah Parks, Yayasan Sabah and 
Universiti Malaysia Sabah for forest access and various kinds 
of support in the field. Field work was made possible and 
supported in the most effective way by warm hospitality and 
assistance from the people and staff at all forest sites. We are 
particularly indebted to Alim Biun, Aloysius Mail, Jadda 
Suhaimi, Jickson Sankin, Suati Selimon and Awang Matamin, 
among many others. Thanks are also due to K. Eduard 
Linsenmair and Brigitte Fiala for logistic support. We are 
grateful to Thomas D. Lambert, Jon Sadler and several 
anonymous reviewers for reading and commenting on drafts 
of the manuscript. Financial support was kindly provided by 
the German Academic Exchange Service (DAAD) to K.W. 
REFERENCES 
Adler, G.H. (2000) Tropical tree diversity, forest structure and 
the demography of a frugivorous rodent, the spiny rat 
(Proechimys semispinosus). Journal of Zoology, 250, 57-74. 
Ashton, P.S. & Hall, P. (1992) Comparisons of structure 
among mixed dipterocarp forests of northwestern Borneo. 
Journal of Ecology, 80, 459-481. 
Asquith, N.M., Wright, S.J. & Clauss, M.J. (1997) Does 
mammal community composition control recruitment in 
neotropical forests? Evidence from Panama. Ecology, 78, 
941-946. 
August, P.V. (1983) The role of habitat complexity and het- 
erogeneity in structuring tropical mammal communities. 
Ecology, 64, 1495-1507. 
Beck, H., Gaines, M.S., Hines, J.E. & Nichols, J.D. (2004) 
Comparative dynamics of small mammal populations in 
treefall gaps and surrounding understorey within Amazo- 
nian rainforest. Oikos, 106, 27-38. 
Bernard, H. (2004) Effects of selective logging on the micro- 
habitat-use patterns of non-volant small mammals in a 
Bornean tropical lowland mixed-dipterocarp forest. Nature 
and Human Activities, 8, 1-11. 
Blate, G.M., Peart, D.R. & Leighton, M. (1998) Post-dis- 
persal predation on isolated seeds: a comparative study of 
40 tree species in a southeast Asian rainforest. Oikos, 82, 
522-538. 
Bourliere, F. (1989) Mammalian species richness in tropical 
rainforests. Vertebrates in complex tropical systems (ed. by L. 
Harmelin-Vivien and F. Bourliere), pp. 153-168. Springer- 
Verlag, New York. 
Brearley, F.Q., Prajadinata, S., Kidd, P.S., Proctor, J. & Sur- 
iantata (2004) Structure and floristics of an old logged rain 
forest in central Kalimantan, Indonesia, and a comparison 
with adjacent primary forest. Forest Ecology and Manage- 
ment, 195, 385-397. 
Brown, J.H., Whitham, T.G., Ernest, S.K.M. & Gehring, C.A. 
(2001) Complex species interactions and the dynamics of 
ecological systems: long-term experiments. Science, 293, 
643-650. 
Cannon, C.H., Peart, D.R. & Leighton, M. (1998) Tree species 
diversity in commercially logged Bornean rainforests. Sci- 
eMce, 281, 1366-1368. 
Chazdon, R.L., Careaga, S., Webb, C. & Vargas, O. (2003) 
Community and phylogenetic structure of reproductive 
traits of woody species in wet tropical forests. Ecological 
Monographs, 73, 331-348. 
Clarke, K.R. (1993) Nonparametric multivariate analyses of 
changes in community structure. Australian Journal of 
Ecology, 18, 117-143. 
Cleary, D.F.R. (2003) An examination of scale of assessment, 
logging and ENSO-induced fires on butterfly diversity in 
Borneo. Oecologia, 135, 313-321. 
10 
2007 The Authors. Journal compilation 
Journal of Biogeography 
2007 Blackwell Publishing Ltd 
Logging impact on Bornean small mammals 
Colon, C.P. (2002) Ranging behaviour and activity of the 
Malay civet (Viverra tangalunga) in a logged and an un- 
tagged forest in Danum Valley, East Malaysia. Journal of 
Zoo/ogy, 257, 473-485. 
Colwell, R.K., Mao, C.X. & Chang, J. (2004) Interpolating, 
extrapolating, and comparing incidence-based species ac- 
cumulation curves. Ecology, 85, 2717-2727'. 
Condit, R., Ashton, P.S., Baker, P., Bunyavejchewin, S., 
Gunatilleke, S., Gunatilleke, N., Hubbell, S.P., Foster, R.B., 
Itoh, A., LaFrankie, J.V., Lee, H.S., Losos, E., Manokaran, 
N., Sukumar, R. & Yamakura, T. (2000) Spatial patterns in 
the distribution of tropical tree species. Science, 288, 1414- 
1418. 
Condit, R., Pitman, N., Leigh, E.G., Chave, J., Terborgh, )., 
Foster, R.B., Nunez, P., Aguilar, S., Valencia, R., Villa, G, 
Muller-Landau, H.C., Losos, E. & Hubbell, S.P. (2002) Beta- 
diversity in tropical forest trees. Science, 295, 666-669. 
Connell, ).H. (1978) Diversity in tropical rain forests and coral 
reefs. Science, 199, 1304-1310. 
Cottingham, K.L., Brown, B.L. & Lennon, J.T. (2001) Biodi- 
versity may regulate the temporal variability of ecological 
systems. Ecological Letters, 4, 72-85. 
Curran, L.M. & Webb, CO. (2000) Experimental tests of the 
spatiotemporal scale of seed predation in mast-fruiting 
Dipterocarpaceae. Ecological Monographs, 70, 129-148. 
Curran, L.M., Trigg, S.N., McDonald, A.K., Astiani, D., 
Hardiono, Y.M., Siregar, P., Caniago, I. & Kasischke, E. 
(2004) Lowland forest loss in protected areas of Indonesian 
Borneo. Science, 303, 1000-1003. 
Davis, A.J., Holloway, J.D., Huijbregts, H., Krikken, )., Kirk- 
Spriggs, A.H. & Button, S.L. (2001) Dung beetles as in- 
dicators of change in the forests of northern Borneo. Journal 
of Applied Ecology, 38, 593-616. 
Denslow, J.S. (1995) Disturbance and diversity in tropical rain 
forests: the density effect. Ecological Applications, 5, 962-968. 
Emmons, L.H. (1991) Frugivory in tree shrews (Tupaia). The 
American Naturalist, 138, 642-649. 
Emmons, L.H. (2000) Tupai: a field study of Bornean tree- 
shrews. University of California Press, Berkeley. 
Ernest, S.K.M. & Brown, J.H. (2001) Delayed compensation 
for missing keystone species by colonization. Science, 292, 
101-104. 
Gascon, G, Lovejoy, T.E., Bierregaard, R.O., Jr, Malcolm, ).R., 
Stouffer, P.C., Vasconcelos, H.L., Laurance, W.F., Zimmer- 
man, B? Tocher, M. & Borges, S. (1999) Matrix habitat and 
species richness in tropical forest remnants. Biological Con- 
servation, 91, 223-229. 
Goodman, S.M. & Rakotondravony, D. (2000) The effects of 
forest fragmentation and isolation on insectivorous small 
mammals (Lipotyphla) on the central high plateau of Ma- 
dagascar. Journal of Zoology, 250, 193-200. 
Gorog, A.)., Sinaga, M.H. & Engstrom, M.D. (2004) Vicariance 
or dispersal? Historical biogeography of three Sunda shelf 
murine rodents {Maxomys surifer, Leopoldamys sabanus and 
Maxomys whiteheadi). Biological Journal of the Linnean So- 
ciety, 81, 91-109. 
Heydon, MJ. & Bulloh, P. (1996) The impact of selective 
logging on sympatric civet species in Borneo. Oryx, 30, 31- 
36. 
Heydon, MJ. & Bulloh, P. (1997) Mouse deer densities in a 
tropical rainforest: the impact of selective logging. Journal of 
Applied Ecology, 34, 484-496. 
Hill, ).K. & Hamer, K.C. (2004) Determining impacts of ha- 
bitat modification on diversity of tropical forest fauna: the 
importance of spatial scale. Journal of Applied Ecology, 41, 
744-754. 
Hewlett, B.E. & Davidson, D.W. (2003) Effects of seed avail- 
ability, site conditions, and herbivory on pioneer recruit- 
ment after logging in Sabah, Malaysia. Forest Ecology and 
Management, 184, 369-383. 
Kitayama, K. (1992) An altitudinal transect study of the 
vegetation on Mount Kinabalu, Borneo. Vegetatio, 102, 149- 
171. 
Laidlaw, R.K. (2000) Effects of habitat disturbance and pro- 
tected areas on mammals of peninsular Malaysia. Conser- 
vation Biology, 14, 1639-1648. 
Lambert, T.D., Adler, G.H., Riveros, CM., Lopez, L., Ascanio, 
R. & Terborgh, J. (2003) Rodents on tropical land-bridge 
islands. Journal of Zoology, 260, 179-187. 
Lambert, T.D., Malcolm, J.R. & Zimmerman, B.L. (2005) Ef- 
fects of mahogany (Swietenia macrophylla) logging on small 
mammal communities, habitat structure, and seed preda- 
tion in the southeastern Amazon basin. Forest Ecology and 
Management, 206, 381-398. 
Lammertink, M. (2004) A multiple-site comparison of 
woodpecker communities in Bornean lowland and hill for- 
ests. Conservation Biology, 18, 746-757. 
Langham, N. (1983) Distribution and ecology of small mam- 
mals in three rain forest localities of peninsula Malaysia with 
particular references to Kedah Peak. Biotropica, 15, 199-206. 
Laurance, W.F. (1994) Rainforest fragmentation and the 
structure of small mammal communities in tropical 
Queensland. Biological Conservation, 69, 23-32. 
Laurance, W.F. & Laurance, S.G.W. (1996) Responses of five 
arboreal marsupials to recent selective logging in tropical 
Australia. Biotropica, 28, 310-322. 
Lawton, J.H., Bignell, D.E., Bolton, B., Bloemers, G.F., Eggle- 
ton, P., Hammond, P.M., Hodda, M., Holt, R.D., Larsen, 
T.B., Mawdsley, N.A., Stork, N.E., Srivastava, D.S. & Watt, 
A.D. (1998) Biodiversity inventories, indicator taxa and ef- 
fects of habitat modification in tropical forest. Nature, 391, 
72-76. 
Lynam, A.J. & Billick, I. (1999) Differential responses of small 
mammals to fragmentation in a Thailand tropical forest. 
Biological Conservation, 91, 191-200. 
MacKinnon, K., Hatta, G, Halim, H. & Mangalik, A. (1996) 
The ecology of Kalimantan - Indonesian Borneo. Periplus 
Editions, Singapore. 
Magurran, A.E. (2004) Measuring biological diversity. Black- 
well, Oxford. 
Malcolm, J.R. (1997) Biomass and diversity of small mammals 
in Amazonian forest fragments.  Tropical forest remnants: 
Journal of Biogeography 
? 2007 The Authors. Journal compilation ? 2007 Blackwell Publishing Ltd 
11 
K. Wells et al. 
ecology, management, and conservation of fragmented com- 
munities (ed. by W.F. Laurance and R.O. Bierregaard Jr), pp. 
207-221. University of Chicago Press, Chicago. 
Malcolm, J.R. & Ray, J.C. (2000) Influence of timber extraction 
routes on central African small-mammal communities, 
forest structure, and tree diversity. Conservation Biology, 14, 
1623-1638. 
Molino, J.F. & Sabatier, D. (2001) Tree diversity in tropical 
rain forests: a validation of the intermediate disturbance 
hypothesis. Science, 294, 1702-1704. 
Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, 
G.A.B. & Kent, J. (2000) Biodiversity hotspots for con- 
servation priorities. Nature, 403, 853-858. 
Newbery, D.M., Campbell, E.J.F., Proctor, J. & Still, M.J. 
(1996) Primary lowland dipterocarp forest at Danum Valley, 
Sabah, Malaysia: species composition and patterns in the 
understorey. Vegetatio, 122, 193-220. 
Ochoa, J. (2000) Effects of logging on small-mammal diversity 
in the lowland forests of the Venezuelan Guyana region. 
Biotropica, 32, 146-164. 
Pardini, R. (2004) Effects of forest fragmentation on small 
mammals in an Atlantic forest landscape. Biodiversity & 
Conservation, 13, 2567-2586. 
Payne, J., Francis, CM. & Phillipps, K. (1998) A field guide to 
the mammals of Borneo. The Sabah Society, WWF Malaysia, 
Kota Kinabalu, Kuala Lumpur. 
Ricketts, T.H., Daily, G.C. & Ehrlich, P.R. (2002) Does 
butterfly diversity predict moth diversity? Testing a pop- 
ular indicator taxon at local scales. Biological Conservation, 
103, 361-370. 
Saiful, A.A., Idris, A.H., Rashid, Y.N., Tamura, N. & Hayashi, 
F. (2001) Home range size of sympatric squirrel species 
inhabiting a lowland dipterocarp forest in Malaysia. Bio- 
tropica, 33, 346-351. 
Schnitzer, SA. & Carson, W.P. (2001) Treefall gaps and the 
maintenance of species diversity in a tropical forest. Ecology, 
82, 913-919. 
Seamon, J.O. & Adler, G.H. (1996) Population performance of 
generalist and specialist rodents along habitat gradients. 
Canadian Journal of Zoology, 74, 1130-1139. 
Sist, P., Shell, D., Kartawinata, K. & Priyadi, H. (2003) Re- 
duced-impact logging in Indonesian Borneo: some results 
confirming the need for new silvicultural prescriptions. 
Forest Ecology & Management, 179, 415-427. 
Sodhi, N.S.A (2002) Comparison of bird communities of two 
fragmented and two continuous southeast Asian rainforests. 
Biodiversity & Conservation, 11, 1105-1119. 
Sodhi, N.S. & Liow, L.H. (2000) Improving conservation 
biology research in southeast Asia. Conservation Biology, 14, 
1211-1212. 
Sodhi, N.S., Koh, L.P., Brook, B.W. & Ng, P.K.L. (2004) 
Southeast Asian biodiversity: an impending disaster. Trends 
in Ecology & Evolution, 19, 654-660. 
StatSoft (2001) Statistica 6.0 for Windows. StatSoft, Hamburg. 
Struhsaker, T.T. (1997) Ecology of an African rainforest: logging 
in Kibale and the conflict between conservation and exploita- 
tion. University Press of Florida, Gainsville. 
Terborgh, J., Lopez, L., Nunez, V.P., Rao, M., Shahabuddin, 
G, Orihuela, G., Riveros, M., Ascanio, R., Adler, G.H., 
Lambert, T.D. & Balbas, L. (2001) Ecological meltdown 
in predator-free forest fragments. Science, 294, 1923-1926. 
Veech, J.A. (2005) Analyzing patterns of species diversity as 
departures from random expectations. Oikos, 108, 149-155. 
Walsh, R.P.D. & Newbery, D.M. (1999) The ecoclimatology of 
Danum, Sabah, in the context of the world's rainforest re- 
gions, with particular reference to the dry periods and their 
impact. Philosophical Transactions of the Royal Society of 
London Series B, Biological Sciences, 354, 1869-1883. 
Wells, K. & Bagchi, R. (2005) Eat in or take away - seed 
predation and removal by rats (Muridae) during a fruiting 
event in a dipterocarp rainforest. Raffles Bulletin of Zoology, 
53, 125-130. 
Wells, K., Pfeiffer, M., Lakim, M.B. & Linsenmair, K.E. (2004) 
Use of arboreal and terrestrial space by a small mammal 
community in a tropical rain forest in Borneo, Malaysia. 
Journal of Biogeography, 31, 641-652. 
Wells, K., Pfeiffer, M., Lakim, M.B. & Kalko, E.K.V. (2006) 
Movement trajectories and habitat partitioning of small 
mammals in logged and unlogged rain forests on Borneo. 
Journal of Animal Ecology, 75, 1212-1223. 
Wu, D.L., Luo, J. & Fox, B.J. (1996) A comparison of ground- 
dwelling small mammal communities in primary and sec- 
ondary tropical rainforests in China. Journal of Tropical 
Ecology, 12, 215-230. 
Yasuda, M., Ishii, N., Okuda, T. & Hussein, N.A. (2003) Small 
mammal community: habitat preference and effects of se- 
lective logging. Pasoh ecology of a lowland rain forest in 
Southeast Asia (ed. by T. Okuda, N. Manokaran, K. Niiya- 
ma, S.C. Thomas and P.S. Ashton), pp. 533-546. Springer- 
Verlag, Tokyo. 
Zubaid, A. & Ariffm, M.K.E. (1997) A comparison of small 
mammal abundance between a primary and disturbed 
lowland rain forest in peninsular Malaysia. Malayan Nature 
Journal, 50, 201-206. 
12 
2007 The Authors. Journal compilation 
Journal of Biogeography 
2007 Blackwell Publishing Ltd 
Logging impact on Bornean small mammals 
BIOSKETCHES 
Konstans Wells conducted this work as part of his PhD at the University of Ulm. His interest lies in the various aspects that 
determine the structure of vertebrate communities and the dynamic performances of the species within them, with an emphasis on 
tropical rain forests and temperate manmade landscapes. 
Elisabeth K. V. Kalko is Professor of Ecology at the University of Ulm with a joint position at the Smithsonian Tropical Research 
Institute. Her interests include tropical and temperate diversity, community ecology and conservation biology, with a particular 
focus on small mammals. 
Maklarin B. Lakim is in charge of the research division of Sabah Parks in Borneo. He deals with the conservation and management 
of ecosystems on Borneo, with a focus on primates and other vertebrates. 
Martin Pfeiffer is a research associate at the University of Ulm. He specializes in biodiversity and macroecology, with particular 
emphasis on ant communities in tropical, desert and temperate ecosystems, as well as tropical small-mammal assemblages. 
Editor: Jon Sadler 
Journal of Biogeography 
? 2007 The Authors. Journal compilation ? 2007 Blackwell Publishing Ltd 
13