Kidney structure and allometry of Argentine desert rodents

Journal of Arid Environments (1999) 41: 453–461
Article No. jare.1998.0472
Available online at http://www.idealibrary.com on

Kidney structure and allometry of Argentine
desert rodents

Gabriela B. Diaz* & Ricardo A. Ojeda

Grupo de Investigaciones de Biodiversidad, Instituto Argentino
de Investigaciones de las Zonas Aridas, CONICET, CC 507,

5500 Mendoza, Argentina

(Received 13 January 1998, accepted 10 November 1998)

Two large groups of rodents inhabit South America: the hystricognaths and the
murids. It has been postulated that while the former show a high degree of
specialization to desert habitats, the murids are not well adapted to xeric
conditions. We studied the renal structure and function of selected desert-
dwelling murid and octodontid (hystricognath) rodents from Argentina to
evaluate levels of adaptation to aridity. Our results show that the murids
Salinomys, Andalgalomys, Calomys and Eligmodontia have the highest renal
indices and urine concentration among Argentine desert rodents. The octo-
dontid Tympanoctomys barrerae shows higher renal indices and urine
osmolarity than those of its close relatives Octomys mimax and Octodontomys
gliroides. We compare the renal traits of the Argentine desert murids with those
of other world-desert rodents such as North American heteromyid rodents.
The results show that these Argentine murid rodents posses renal adaptations
to conditions which are as impressive as those of both octodontids and the
classic desert-adapted heteromyids.

1999 Academic Press

Keywords: renal morphology; arid habitats; murids; octodontids; body size

Introduction

Mammalian kidneys control the concentration of the body fluids. Some arid-adapted
small mammals are efficient at concentrating urine in order to reduce water loss.
Their unilobular kidneys have elongated renal papillae, i.e. thicker medulla (Sperber,
1944; McMillen & Lee, 1969; Geluso, 1978; Brooker & Withers, 1994). According to
the countercurrent multiplier model, the ability to concentrate urine depends on the
length of the loops of Henle and collecting ducts that traverse the renal medulla
(Schmidt-Nielsen et al., 1961). The maximum length of the loop of Henle is directly
proportional to medullary thickness (Beuchat, 1990). Medullary thickness has been
used to propose several indices of renal performance (Sperber, 1944; Brownfield
& Wunder, 1976; Geluso, 1978). Higher indices are typical of desert small rodents.
However, other morphological characteristics besides maximum nephron length af-
fect the urinary concentrating ability (Bankir & Rouffignac, 1985).

*(E-mail: [email protected]). 1999 Academic Press
0140}1963/99/040453#09 $30.00/0

454 G. B. DIAZ & R. A. OJEDA

Most variations in gross renal structure in mammals can be explained by variations in
body mass, but it is also influenced by habitat and food habits (Beuchat, 1996). The
ability to concentrate urine is inversely related to body mass (Calder, 1984). Absolute
medullary thickness increases with increasing body mass, and conversely medullary
thickness (RMT) decreases with increasing body mass (Greegor, 1975; Beuchat, 1990),
suggesting that, in general, small species have more powerfully-concentrating kidneys
than do large species. The RMT’s of mammals living in arid areas are greater than those
of similar-sized mammals from more mesic areas (Sperber, 1944). Blake (1977) found
this relation to also be true at the intraspecific level.

Renal morphology and allometry have been studied in different taxa of small
desert mammals, including Australian marsupials, and North American heteromyid and
sciurid rodents (Blake, 1977; Lawler & Geluso, 1986; Brooker & Withers, 1994). There
are few studies of small to medium-sized mammals from xeric habitats of South America
(Cortes et al., 1990; Greegor, 1975).

The South American desert rodent fauna is composed of two major lineages: the
hystricognaths and the murids. Current wisdom regarding desert rodent adaptations has
it that murids are not highly specialized for life in xeric habitats due to their relatively
recent colonization of these habitats (Mares, 1975, 1976), assuming colonization of the
continent after the completion of the Panama land bridge (c. 3·5 million years ago;
Marshall et al., 1979). Data on octodontid (hystricognath) rodents, on the other hand,
indicate a high degree of desert specialization (Mares et al., 1997; Ojeda et al., 1996).
Octodontids have been associated with desert habitats since the Oligocene (Patterson
& Pascual, 1972), and therefore probably far longer periods than have South American
murids.

In the present study, variations in the renal structure of 13 species of desert murids
and octodontid (hystricognath) rodents from Argentina are analysed. In particular, we
analyse the allometry of renal structure and compare it to other groups of mammals,
especially rodents from other deserts.

Materials and methods

Thirteen rodent species (four octodontids and nine murids) were captured using
Sherman live traps and Museum Special snap traps in several arid and semi-arid sites of
Argentina in the lowland Monte Desert, the highland desert or Puna, and the semi-arid
thornscrub or Chaco (Table 1). All specimens were preserved as skins, skulls and
skeletons and housed in the Instituto Argentino de Investigaciones de las Zonas Aridas
(IADIZA) Mammal Collection, Mendoza, Argentina. Most species are herbivorous,
with the exception of the grass mouse Akodon molinae that incorporates approximately
30% of arthropods in its diet (Campos, 1997).

Anatomical procedures

Both kidneys from adult animals were removed within 24 h of death and fixed in 10%
formaldehyde. Length, width and thickness of the kidneys were measured with a dial
caliper to the nearest 0·5 mm. Sections along the frontal axis through the longest part of
the renal papilla were obtained from the left kidney, but if a complete saggital section
could not be made, the right kidney was used. Thickness of the cortex (CT), outer zone
of the medulla and inner zone of the medulla were distinguishable when viewed through
a stereoscopic microscope (6;); medulla thickness (MT) is the sum of the values of
outer and inner medulla. The outline of the entire kidney, including the corticomedul-
lary junction, and the outer and inner boundaries of the medulla were traced on paper
using a camera lucida attached to a stereoscopic microscope (Geluso, 1978). The

Table 1. Species of Argentina desert rodents examined: collecting localities, coordinates, biomes and precipitation KIDNEY STRUCTURE OF ARGENTINE DESERT RODENTS

Species Collecting localities Coordinates Biomes Rainfall range (mm)

Muridae N acun an, Mendoza 34302 S, 67358 W Monte 300}400 mm
Amanao, Catamarca 27333 S, 66331 W Monte 200}300 mm
Akodon molinae La Antigua, La Rioja 30302 S, 66304 W Chaco 300}400 mm
Andalgalomys olrogi N acun an, Mendoza 34302 S, 67358 W Monte 300}400 mm
Andalgalomys roigi Talampaya, La Rioja 29347 S, 67354 W Monte 100}200 mm
Calomys musculinus N acun an, Mendoza 34302 S, 67358 W Monte 300}400 mm
Eligmodontia moreni N acun an, Mendoza 34302 S, 67358 W Monte 300}400 mm
Eligmodontia typus Talampaya, La Rioja 29347 S, 67354 W Monte 100}200 mm
Graomys griseoflavus La Antigua, La Rioja 30302 S, 66304 W Chaco 300}400 mm
Phyllotis xanthopygus
Salinomys delicatus N acun an, Mendoza 34302 S, 67358 W Monte 300}400 mm
Ischigualasto, San Juan 29347 S, 67354 W Monte 100}200 mm
Octodontidae Saladillo, Jujuy 23343 S, 65306 W Puna 200}300 mm
Nihuil, Mendoza 35302 S, 68340 W Monte 200}300 mm
Ctenomys eremophilus
Octomys mimax
Octodontomys gliroides
Tympanoctomys barrerae

455

456 G. B. DIAZ & R. A. OJEDA

thickness of each zone was measured with a ruler (0·5 mm). The area of each zone was
measured with a LI-COR 3000A portable area meter. The following indices were
calculated: relative medullary thickness (RMT"medullary thickness;10/cube root of
the product of kidney length, width and thickness); ratio of inner medulla to cortex
(MI/C); and relative medullary area (RMA"medullary area/cortical area) (Sperber,
1944; Brownfield & Wunder, 1976; Geluso, 1978).

Urine concentration

Urine samples of Tympanoctomys barrerae (N"9, 51 samples), Octomys mimax (N"7,
18 samples), Eligmodontia typus (N"16, 33 samples), Calomys musculinus (N"5, 18
samples) and Salinomys delicatus (N"1, 2 samples) were collected in Petri dishes
placed under the wire-grid floor of the animals’ cages and covered with a layer of mineral
oil to prevent evaporation. After 24 h the urine was pipetted into capped plastic vials,
avoiding samples contaminated by faeces. The osmolarity of the urine (mosm/l) was
determined on a vapour pressure osmometer (Wescor) using dilutions in distilled water.
Maximum values were obtained from the field in the case of octodontids and from the
laboratory in the case of murids (maintained on a diet of sunflower seeds).

Statistical analysis

Relationships between log-transformed data of body mass (M) and renal characteristics
(CT, MT, RMT, MI/C and RMA) were assessed by Pearson’s correlation analysis.
Least-squares regression analysis was performed for all the Argentine desert rodents.
The allometric equations are of the form log y"log a#b log M (M"body mass in g).
Comparisons among slopes and intercepts were made with the t test (Zar, 1984).
Statistical analyses were performed using STATISTICA program version 5.0.

Results

Renal morphology

Kidneys of the Argentine desert rodents studied show a single renal papilla and
a medulla divided in two zones (Fig. 1). The delicate mouse, Salinomys delicatus, has the
most elongated papillae and the highest renal indices (MI/C, RMT and RMA) among
the murids in this sample. The same is true for the red vizcacha rat, Tympanoctomys
barrerae, among the octodontids (Table 2). On the other hand, shorter papillae are
associated with lower renal indices, as in Akodon molinae (murid) and Octomys mimax
(octodontid; Fig. 1).

Allometry

Pearson’s correlation coefficients between mean cortex thickness, renal indices
(RMT, MI/C and RMA) and mean body mass of Argentine desert rodents (murids and
octodontids) were consistently significant (Table 3). Thus, smaller species have a thin-
ner cortex and greater indices, but not more developed medullary thickness compared to
larger rodents (Table 3).

No significant correlation was found between renal morphological characteristics of
octodontids and body mass. Renal indices of Argentine murids correlates significantly
with body mass (MI/C: R"0·86, N"9, p"0·029; RMA: R"0·708, N"9,

KIDNEY STRUCTURE OF ARGENTINE DESERT RODENTS 457

Figure 1. Renal morphology of four selected murid (Salinomys delicatus, Akodon molinae) and
octodontid rodent species (Tympanoctomys barrerae, Octomys mimax). C"cortex; M"outer
medulla; Mi"inner medulla.

p"0·033; RMT: R"0·656, N"9, p"0·055). The RMT vs. M regression line for

Argentine murid (log RMT"1·289!0·167 log M; R"0·656, N"9, p"0·055) is

used to compare with North American heteromyids (log RMT"1·195!0·1695 log

M; R"0·976, N"6, p"0·0015) (Lawler & Geluso, 1986). Slopes did not differ

Tsighnisifiicnadnictlayte(sbAt"hat!th0e·1lin7,esp'of A0·r0g5e)ntbinuet intercepts did (t"1·99, df."76, p(0·05).
murids and heteromyids are parallel but not

coincident. Therefore, Argentine desert murids show higher RMT than do desert

heteromyids.

Urine concentration
Maximum values of urine osmolarity are shown in Table 4.

Discussion

The high renal indices and maximum urine osmolarity of Calomys musculinus and
Eligmodontia typus (Tables 2 and 4) correspond with the physiological ability of Calomys
musculinus to survive on a diet of seeds and the ability of Eligmodontia typus to tolerate
highly concentrated (2)00 M) solutions of salt (Mares, 1977a,b). The high renal indices
of Salinomys delicatus, Andalgalomys roigi and Andalgalomys olrogi show that the kidneys
of these desert murids are also highly specialized for desert existence. Moreover, their
renal indices and maximum urine osmolarity are among the highest values reported for
other desert rodents (Table 4; Fig. 2).

Among octodontids, the red vizcacha rat, Tympanoctomys barrerae possesses a kidney
with a long broad renal papilla and produces highly concentrated urine. These renal
attributes are similar to those of the fat sand rat Psammomys obesus (Table 4; Fig. 2) that
excretes a large amount of fluid with a high concentration of salt (Schmidt-Nielsen,
1961). Both species have a specialized diet of halophytic chenopods and live in salt
dunes habitats (Ojeda et al., 1996; Mares et al., 1997). Close relatives of Tympanoctomys
barrerae, such as Octomys mimax and Octodontomys gliroides, have less-developed renal

Table 2. Renal indices ( $SE) for the desert rodents of Argentina. Number of individuals are indicated in parentheses when the numbers for 458 G. B. DIAZ & R. A. OJEDA
MI/C or RMA were different

Species N M (g) CT (mm) MT (mm) RMT MI/C RMA

Muridae 1 12·50 0·83 7·83 13·98 7·02 1·69
11(6) 12·73$0·27 0·82$0·02 6·94$0·04 12·29$0·10 6·58$0·38 1·47$0·04
Salinomys delicatus 14(10) 14·50$0·13 0·99$0·01 6·72$0·12 10·66$0·21 5·32$0·12 1·63$0·06
Calomys musculinus 12 (10) 18·58$0·25 0·80$0·01 6·97$0·05 11·42$0·11 6·36$0·08 1·63$0·04
Eligmodontia moreni 1 23·00 1·17 8·66 13·48 5·84 1·18
Eligmodontia typus 1 28·00 1·33 8·66 12·60 4·01 1·06
Andalgalomys olrogi 14 (7) 31·11$0·35 1·17$0·03 7·01$0·05 10·09$0·08 4·27$0·22 1·44$0·04
Andalgalomys roigi 3 44·00$4·67 1·63$0·02 8·44$0·36 10·10$0·47 3·68$0·19 0·96$0·07
Akodon molinae 28 (25) 54·50$0·37 1·44$0·01 8·72$0·03 4·24$0·04 1·27$0·01
Phyllotis xanthopygus 9·64$0·04
Graomys griseoflavus 16 (12) 86·79$1·00 1·72$0·01 1·73$0·03
11 (6) 98·32$1·57 1·74$0·02 12·38$0·07 9·41$0·07 5·00$0·06 0·83$0·01
Octodontidae 14 (11) 121·18$2·26 1·97$0·04 6·38$0·08 6·09$0·06 2·58$0·03 0·75$0·02
1 153·00 1·83 9·61$0·07 8·98$0·08 3·91$0·07 1·03
Tympanoctomys barrerae 6·83 5·35 3·19
Octomys mimax
Ctenomys eremophilus
Octodontomys gliroides

M"body mass; CT"cortex thickness; MT"medullary thickness; RMT"relative medullary thickness; MI/C"ration of inner medulla cortex;
RMA"relative medullary area.

KIDNEY STRUCTURE OF ARGENTINE DESERT RODENTS 459

Table 3. Allometry of renal morphology of the 13 species of desert rodents from Argentina
a b RFp

CT !0)44$0)061 0)347$0)038 0)94 84)036 (0)0001
MT 0)80$0)098 0)063$0)061 0)30 1)0884 0)319
RMT 1)425$0)087 !0)269$0)054 0)83 0)0004
MI/C !0)273$0)061 0)80 25)138 0)0009
RMA 1)093$0)099 !0)207$0)073 0)65 20)172 0)0167
0)420$0)119
7)946

CT"cortex thickness; MT"medullary thickness; RMT"relative medullary thickness; MI/C"ratio
inner medulla/cortex; RMA"relative medullary area.

papilla (Fig. 2), low renal indices and low urine concentration (Table 4). This may
reflect a water-rich diet of cacti, as found in the wood rat Neotoma lepida (Stallone, 1979;
Ojeda et al., 1999).

The pattern of decreasing renal indexes and an increase in cortex thickness with
increasing body mass in Argentine desert rodents agrees with previous studies for
different taxa of mammals (Greegor, 1975; Lawler & Geluso, 1986; Brooker
& Withers, 1994). However, Argentine desert rodents show higher slopes than mam-
mals of xeric habitats (Beuchat, 1996). They also do not show a relatively more
developed medullary thickness compared to larger rodents, as has been reported for
mammals (Beuchat, 1990). The smaller range of body mass considered in our study
may account for these differences.

Compared with heteromyids, Argentine murids exhibit higher RMT and metabolic
rate (Bozinovic & Rosenmann, 1988; Bozinovic, 1992). This may explain the dif-
ferent intercept values for murids and heteromyids in the regression line. Granivorous
heteromyids have low metabolic rate and low evaporative water loss, thereby facilitating
water conservation (Hinds & MacMillen, 1985). Higher metabolic rates in Argentine
desert rodents result in greater respiratory exchange and concomitant increases in water
loss. More efficient kidneys (i.e. higher RMT) may compensate for high evaporative
water loss.

Table 4. Maximum urine osmolarity (mosm/l) of desert rodents

Species Maximum urine Source
osmolarity
(mosm/1)

Muridae 8773 This study
8612 This study
Calomys musculinus 7440 This study
Eligmodontia typus 9370 MacMillen & Lee, 1967
Salinomys delicatus 6340 Schmidt-Nielsen, 1964
Notomys alexis
Psammomys obesus 7080 This study
2071 This study
Octodontidae
7500 Altschuler et al., 1979
Tympanoctomys barrerae 5540 Schmidt-Nielsen, 1964
Octomys mimax

Heteromyidae

Perognathus penicillatus
Dipodomys merriami

460 G. B. DIAZ & R. A. OJEDA

Figure 2. Relationship between mean relative medullary thickness, and mean body mass for
Argentine desert rodents (this study, solid line) and similar sized non-Argentine desert rodents.
(᭹)"Argentine Muridae: Sd"Salinomys delicatus; Cm"Calomys musculinus; Em"Elig-
modontia moreni; Et"Eligmodontia typus; Ao"Andalgalomys olrogi; Ar"Andalgalomys roigi;
Am"Akodon molinae; Px"Phyllotis xanthopygus; Gg"Graomys griseoflavus; (᭺)"Octodon-
tidae: Tb"Tympanoctomys barrerae; Om"Octomys mimax; Ce"Ctenomys eremophilus;
Og"Octodontomys gliroides; (᭿)"Non-Argentina Muridae: Na"Notomys alexis;
Nc"Notomys cervinus (MacMillen & Lee, 1969); Ps"Psammomys obesus (Sperber, 1944);
(᭡)"Heteromyidae, Dm"Dipodomys merriami; Mp"Microdipodops pallidus; Pi"Perognathus
longimembris (Lawler & Geluso, 1986).

In conclusion, the renal morphology and function of Argentine desert rodents, either
murids or octodontids, evidence the capacity for reducing water loss. Moreover, it is
suggested that these renal specializations to arid habitats are independent of the time of
South American rodent colonization.

The authors thank our fellow colleagues of the Biodiversity Research Group, C. Campos and
C. Borghi, for their critical comments, suggestions and statistical advice on earlier drafts of the
manuscript. We also thank Drs P. Hanford and M. Willig. Ana M. Scollo kindly helped with the
extraction of kidneys, and M. Gallardo kindly provided some animals. We thank Dra P. Valles,
Dept. Physiopathology, University of Cuyo, for the loan of the vapour pressure osmometer
Wescor 5500, and Ing. Cavagnaro, Plant Physiology Dept., University of Cuyo, for the loan of the
area meter. We thank D. Duen as and R. Marti (Magraf) for helping with figures. This project was
partially funded by the National Council of Scientific and Technical Investigations (CONICET,
Argentina) grants PID 3363800 and PIP 4684.

References

Altschuler, E.M., Nagle, R.B., Braun, E.J., Lindstedt, S.L. & Krutzsch, P.H. (1979). Morpho-
logical study of desert heteromyid kidney with emphasis on the genus Perognathus. Anatomical
Record, 194: 461}468.

Bankir, L. & Rouffignac, C. (1985). Urinary concentrating ability: insights from comparative
anatomy. American Journal of Physiology, 249: R643}R666.

Beuchat, C.A. (1990). Body size, medullary thickness, and urine concentrating ability in mam-
mals. American Journal of Physiology, 258: R298}R308.

Beuchat, C.A. (1996). Structure and concentrating ability of the mammalian kidney: correlations
with habitat. Americal Journal of Physiology, 271: R157}R179.

KIDNEY STRUCTURE OF ARGENTINE DESERT RODENTS 461

Blake, B.H. (1977). The effects of kidney structure and annual cycle on water requirements
in golden-mantled ground squirrels and chipmunks. Comparative Biochemistry and Physiology,
58A: 413}419.

Bozinovic, F. (1992). Rate of metabolism of grazing rodents from different habitats. Journal
of Mammalogy, 73: 379}384.

Bozinovic, F. & Rosenmann, M. (1988). Comparative bioenergetics of South American cricetid
rodents. Comparative Biochemistry and Physiology, 91A: 195}202.

Brooker, B. & Withers, P. (1994). Kidney structure and renal indices of dasyurid marsupials.
Australian Journal of Zoology, 42: 163}176.

Brownfield, M.S. & Wunder, B.A. (1976). Relative medullary area: a new structural index for
estimating urinary concentrating capacity of mammals. Comparative Biochemistry and Physi-
ology, 55: 69}75.

Calder, W.A.III. (1984). Size, Function, and Life History. Cambridge: Harvard University Press.
431 pp.

Campos, C. (1997). UtilizacioH n de recursos alimetarios por mammH feros medianos y pequen os del
desierto del Monte. Doctoral thesis, Universidad Nacional de CoH rdoba. 183 pp.

CorteH s, A., Rosenmann, M. & Baez, C. (1990). FuncioH n del rin oH n y del pasaje nasal en la
conservacioH n de agua corporal en roedores simpaH tridos de Chile central. Revista Chilena de
Historia Natural, 63: 279}291.

Geluso, K.N. (1978). Urine concentrating ability and renal structure of insectivorous bats. Journal
of Mammalogy, 59: 312}323.

Greegor, D.H. (1975). Renal capabilities of an Argentine desert armadillo. Journal of Mammalogy,
56: 626}632.

Hinds, D.S. & MacMillen, R.E. (1985). Scaling of energy metabolism and evaporative water loss
in heteromyid rodents. Physiological Zoology, 58: 282}298.

Lawler, R.M. & Geluso, K.N. (1986). Renal structure and body size in heteromyid rodents.
Journal of Mammalogy, 67: 367}372.

MacMillen, R.E. & Lee, A.K. (1967). Australian desert mice: independence of exogenous water.
Science, 158: 383}385.

MacMillen, R.E. & Lee, A.K. (1969). Water metabolism of Australian hopping mice. Compara-
tive Biochemistry and Physiology, 28: 493}514.

Mares, M.A. (1975). South American mammal zoogeography: evidence from convergent evolu-
tion in desert rodents. Proceedings National Academy of Sciences, 72: 1702}1706.

Mares, M.A. (1976). Convergent evolution of desert rodents: multivariate analysis and zoogeo-
graphic implications. Paleobiology, 2: 39}63.

Mares, M.A. (1977a). Water economy and salt balance in a South American desert rodent,
Eligmodontia typus. Comparative Biochemistry and Physiology, 56A: 325}332.

Mares, M.A. (1977b). Water independence in a South American non-desert rodent. Journal of
Mammalogy, 58: 653}656.

Mares, M.A., Ojeda, R.A., Borghi, C.E., Giannoni, S., Diaz, G.B. & Braun, J.K. (1997). A desert
rodent uses hair as a tool to overcome halophytic plant defenses. BioScience, 47: 699}704.

Marshall, L.G., Butler, R.F., Drake, R.E., Curtis, G.G. & Tedford, R.H. (1979). Calibration of
the Great American Interchange. Science, 204: 272}279.

Ojeda, R.A., Gonnet, J., Borghi, C.E., Giannoni, S.M., Campos, C. & Diaz, G.B. (1996).
Ecological observations of the red vizcacha rat Tympanoctomys barrerae in two desert habitats of
Argentina. Mastozoologn&a Neotropical, 3: 183}191.

Ojeda, R.A., Borghi, C.E., Diaz, G.B., Giannoni, S.M., Mares, M.A. & Braun, J.K. (1999).
Evolutionary convergence of the highly adapted desert rodent Tympanoctomys barrerae (Octo-
dontidae). Journal of Arid Environments, 41: 443}452.

Patterson, B. & Pascual, R. (1972). The fossil mammal fauna of South America. In: Keast, A.,
Erk, E. & Glass, B. (Eds), Evolution, mammals and Southern continents, pp. 247}309. Albany
Publications, State University New York Press.

Schmidt-Nielsen, B., O’Dell, R. & Osaki, H. (1961). Structure and concentrating mechanism in
the mammalian kidney. American Journal of Physiology, 200: 1119}1124.

Schmidt-Nielsen, K. (1964). Desert animals: physiological problems of heat and water. London:
Oxford University Press. 277 pp.

Sperber, I. (1944). Studies on the mammalian kidney. Zoologiska Bidrag fran Uppsala, 22: 249}431.
Stallone, J.N. (1979). Seasonal changes in water metabolism of woodrats. Oecologia, 38: 203}216.
Zar, J.H. (1984). Biostatistical analysis. Second edition. Prentice-Hall International, Inc. 718 pp.