INTEG.
AND
COMP. BIOL., 42:43–50 (2002)
Reproductive Effort and Reproductive Nutrition of Female Desert Tortoises:
Essential Field Methods1
BRIAN THOMAS HENEN2
Department of Biology, University of California, Los Angeles, California 90095-1606
SYNOPSIS.
I used three innovative, nondestructive field methods (gas dilution, doubly labeled water and
radiography) to measure individual energy and water budgets of wild, female desert tortoises (Gopherus
agassizii). With these budgets, I evaluated whether body reserves help females produce eggs independent of
rainfall and food availability. Female desert tortoises used large seasonal and annual changes in metabolism
and body water, protein and energy reserves to survive and produce eggs. Although lipid reserves are
important to female desert tortoises, nitrogen or crude protein appears to be the primary limiting resource
for producing eggs. By reducing metabolic rates 90%, females conserved enough body reserves to produce
eggs during extreme drought conditions; this is an effective bet-hedging reproductive pattern in an extreme
and unpredictable environment.
INTRODUCTION
Physiological ecologists often pursue field studies in
harsh environments in search of extreme physiological
tolerances, adaptations or exaptations enabling organisms to endure in these environments. Consequently,
the extremely low, variable and unpredictable rainfall
and primary productivity of deserts (Pianka, 1978;
Louw and Seely, 1982; Henen, 1997; Henen et al.,
1998) has intrigued many physiological ecologists
(Louw and Seely, 1982; Schmidt-Nielsen, 1997).
Rainfall and primary or secondary productivity are
strong cues for reproduction in desert organisms, yet
reproductive strategies of desert animals are diverse
(e.g., Kenagy and Bartholomew, 1985; Turner et al.,
1984a, b). Despite experiencing a 10-fold variation in
their food availability, desert tortoises (Gopherus
agassizii) did not vary their annual egg production in
1983–85 (Henen, 1997). How did female desert tortoises reproduce seemingly independent of rainfall and
food availability?
To produce eggs independent of rainfall and food
availability, female desert tortoises may store energy
before winter and use these reserves the ensuing spring
to produce eggs. To evaluate this possibility, I measured annual energy budgets of female G. agassizii.
These budgets were also designed to quantify reproductive effort (Congdon et al., 1982; Nagy, 1983) and
identify periods when nutrient limitations influence female activity and reproduction. I used three nondestructive methods to measure energy allocations to
field metabolism (doubly labeled water), eggs (X-ray
radiography) and body lipid and non-lipid energy content (gas dilution). Gas dilution requires a long equilibration and measuring many equilibration parameters.
However, these measurements enable accurate, nondestructive estimation of lipid mass (Lesser et al.,
1952; Henen, 1991, 2001; Gessaman et al., 1998) and
nonlipid mass (Henen, 1991, 2001). Gas dilution is
accurate for animals with low, or slowly changing, lipid reserves, and variable body temperatures and hydration states (e.g., desert tortoises).
This analysis focuses on field evaluations of the
body composition of wild, female desert tortoises. In
doing so, it demonstrates physiological, ecological and
evolutionary features enabling desert tortoises to survive and reproduce for decades in the harsh, variable
and unpredictable desert clime.
MATERIALS AND METHODS
Study design
The study site was in remote creosote bush (Larrea
tridentata)—burrobush (Ambrosia dumosa) scrub habitat of the eastern Mojave Desert (348519N, 1158099E,
680 m elevation). Female G. agassizii become active
in early spring (ca. mid-March to mid-April) and produce eggs in spring (ca. mid-April to mid-July). The
tortoises use burrows to avoid the hot midday hours
in summer (July to September) and enter winter burrows in autumn (ca. mid-October), remaining dormant
until March. To evaluate budgets for two reproductive
cycles, I recaptured nine female G. agassizii, fitted
with radiotransmitters, from July 1987 to July 1989.
Females were radiographed every two weeks during
each spring to quantify egg production. Body composition and doubly labeled water data were measured
at least once each season (Fig. 1).
Doubly labeled water and X-ray radiography
The doubly labeled water and X-ray radiography
methods were detailed in Henen (1997) and will only
be briefly outlined. At the beginning of the study, a
100 ml brachial or supraorbital blood sample was collected from each female. Females were then injected
intramuscularly with a calibrated volume of doubly labeled water (oxygen-18 and tritium or deuterium). After $14 hour of equilibration, another small blood
1 From the Symposium Taking Physiology to the Field: Advances
in Investigating Physiological Function in Free-Living Vertebrates
presented at the Annual Meeting of the Society for Integrative and
Comparative Biology, 3–7 January 2001, at Chicago, Illinois.
2 Present address of Brian T. Henen is Department of Zoology,
University of the Western Cape, Private Bag X17, Bellville 7535,
South Africa; E-mail: [email protected]
43
44
BRIAN THOMAS HENEN
licles before ovulation (Rostal et al., 1994; B. T. Henen
and O. T. Oftedal, unpublished data). After ovulation
however, females add albumen and shell to complete
each egg and proceed to lay their eggs. Because of this
commitment at ovulation, I assigned reproductive allocations to spring periods when females ovulated.
FIG. 1. Mean and 95% CI of body mass (Mb), total body water
(Wb) and dry matter (DM; all in g), plus Wb as % of Mb, of nine
female desert tortoises for two years. Clutch mass is illustrated for
comparative purposes and rainfall (60.1 mm) was measured on site
about once per month. In March and April 1989, n 5 7 and 8,
respectively. The body measures changed through time (RMANOVA F9,86 . 25, 26, 4 and 5 and P , 10217, 10218, 1023, 1024, respectively). Means without adjacent numerals were the lowest means
or not significantly higher than the lowest means. The numeral labeling a mean represents the number of lowest means exceeded by
that mean (e.g., for body mass, April 1988 . eight of the other nine
means; determined by Student-Newman-Keuls multiple comparisons).
sample was collected for determining equilibrium isotope concentration and total body water (in g) by isotopic dilution. I then released the animals to range
freely until recaptured.
Upon recapture of each animal, I collected voided
urine or 100 ml of blood, injected 0.9 ml of singly
labeled water (tritiated or deuterated water) or doubly
labeled water, collected blood after isotopic equilibration, and released the animal. Reinjections maintained
isotope levels for accurate analysis and helped measure
total body water for body composition calculations.
Sample isotope concentrations were used to calculate
field metabolic rates (liter CO2 produced/day, kJ/day
and MJ/yr).
Females were radiographed using a portable X-ray
machine (MinXray 300) and a technique of 70 kVP
and 5 mA-sec. Eggshells become radio-opaque within
days after ovulation (unpublished data, B.T.H.), allowing quantification of clutch size, egg size and annual
fecundity (Turner et al., 1986; Henen, 1994, 1997;
Wallis et al., 1999). Using fecundity and the energy
content of captive-tortoise eggs, I calculated the chemical potential energy each female invested in eggs. Ultrasonography indicates that females can reabsorb fol-
Body composition
The body composition of each female was measured
10 times (with two exceptions; Fig. 1) using the dilution of cyclopropane gas; consult Henen (1991,
1997, 2001) for details. An organism inside a glass
chamber with a lipid soluble gas (e.g., cyclopropane)
will absorb much of the gas in its lipids. For cyclopropane dilutions on female desert tortoises, I measured the cyclopropane concentration of chamber gas
samples in the field on a portable gas chromatograph
(Foxboro OVA128BIE-GC Gas Analyzer). The moles
of cyclopropane absorbed in each female’s lipids were
calculated using the 1) concentration of cyclopropane
in chamber gas samples, 2) solubilities of cyclopropane for nonlipid matter, 3) temperature, volume and
gas pressure inside the chamber, and 4) animal’s body
temperature, approximate volume, and total body water. I then used the moles of cyclopropane in the lipids
and the lipid solubility coefficient to calculate body
lipid mass (in g).
I calculated body dry mass (5body mass 2 total
body water) and nonlipid dry mass (5dry mass 2 lipid
mass) by difference. Lipid and nonlipid dry masses
were converted to energy units (kJ or MJ) using energy densities (37.5 and 9.80 kJ/g, respectively) measured with bomb calorimetry on lipid and nonlipid extracts of roadkill carcasses. For each period I calculated individual anabolic and catabolic rates (positive
and negative values, respectively; kJ/day or MJ/yr)
from serial measurements of body lipid and nonlipid
energy. Metabolizable energy intake (kJ/day or MJ/yr)
equaled the sum of field metabolic rate, chemical potential energy of eggs produced, and the change in
body energy content (5lipid 1 nonlipid energy).
Statistics
Body composition measures were summarized as
means 695% CI and compared among periods using
one-way repeated measures ANOVA (RMANOVA, all
df 5 9 and 86). RMANOVA were followed by Student-Newman-Keuls multiple comparisons (SigmaStat
2.0 for Windows, Jandel Scientific) to determine which
means differed. Results were significant at P , 0.05.
RESULTS AND DISCUSSION
Egg production was lower in 1988 and 1989 than
in 1983–1985 (Henen, 1993, 1997), so G. agassizii
reproduction is affected by rainfall and food availability (see also Turner et al., 1987; Wallis et al., 1999).
The relationship of egg production to food abundance
is not simple, egg production appears to approach an
asymptote at moderate to high food abundances, like
those in 1983–1985 (Henen, 1993; Wallis et al., 1999).
TORTOISE REPRODUCTIVE NUTRITION
However, females were able to produce a few eggs in
1988 and 1989 by drawing upon body nutrient reserves (Henen, 1994, 1997).
Body mass, water, and dry matter
Body mass and body water changed significantly
through time (Fig. 1) and showed more extreme
changes than did dry matter and percent body water.
Body mass and body water were highly correlated (r
5 0.99) and increased substantially when females
drank rainwater (both summers and spring 1988) or
ate fresh annual plants or succulent perennial plants
(early spring, both years; Henen, 1994). Body mass
and body water decreased in spring 1988 and 1989
primarily because egg production was demanding (Henen, 1994, 1997; Wallis et al., 1999).
Body mass and body water increased 7 and 12%,
respectively, during the first year, which included
many opportunities for females to drink rainwater and
consume fresh annual plants. However, the second
year included a very dry winter followed by very low
plant productivity. Consequently, three small females
forfeited egg production in 1989, posting modest gains
in body mass and water (9 and 5%, respectively) and
an 18% increase in dry matter. Conversely, the other
six females produced about four eggs each, and suffered 12, 15 and 6% losses in body mass, water and
dry matter, respectively (Henen, 1994), emphasizing
the importance of body reserves to egg production.
Changes in body mass and body water exceeded
those for body dry matter on an absolute scale (Fig.
1), but less so on a relative scale (Table 1). Dry matter
peaked when tortoises ate fresh food (Summer 1988;
early Spring 1988 and 1989; Henen, 1997) but declined each spring (1988: 72 g, 1989: 64 g), in part
due to substantial allocations of dry matter to eggs (33
and 32 g, respectively). Dry matter did not change
annually except during the dry year, declining 30 g for
egg producing females, which invested 47 g of dry
matter in eggs, and increasing 63 g for females forgoing reproduction (Henen, 1994).
The large changes in body composition are consistent with G. agassizii’s behavioral (Medica et al.,
1980; Nagy and Medica, 1986; Henen, 1994), morphological (Schmidt-Nielsen, 1997) and physiological
features (Minnich, 1982; Nagy and Medica, 1986; Henen, 1994, 1997; Peterson, 1996; Henen et al., 1998;
Christopher et al., 1999) for acquiring water and reducing water losses. Desert tortoises can drink as much
as 40 or 50% of their body mass in water within 15
min (Henen, 1994; unpublished data, B.T.H.) ranking
them with camels (Schmidt-Nielsen, 1997) and amphibians (Pough et al., 1998) in rehydration abilities.
The dehydration tolerances of desert tortoises are also
exceptional (Nagy and Medica, 1986; Henen, 1994;
Peterson, 1996).
Lipid metabolism
Changes in dry matter reflected primarily the changes in nonlipid dry matter (Fig. 2). A simple compari-
45
son of RMANOVA F values, P values and the number
of significant differences among means (Figs. 1 and
2), suggests that dry matter varied less in time than
did lipid mass and nonlipid dry mass. Similar analyses
suggest that lipid mass, and lipid mass relative to body
mass (Fig. 2) or lipid mass relative to nonlipid dry
matter (F9,86 5 43.5, P 5 5.50 3 10225), varied more
in time than did nonlipid dry matter. This is consistent
with perceptions that lipids are more labile than nonlipids (Pond, 1978; Henen, 1991, 1994, 1997).
Body nonlipid energy and lipid energy were simple
products of nonlipid and lipid masses, respectively,
and thus had the same statistical results as did nonlipid
and lipid masses (Figs. 2 and 3). Since lipid holds
more energy per gram than does nonlipid dry matter
(e.g., 38 and 9.8 kJ/g, respectively), small errors in
lipid mass estimates (Fig. 2) can affect the accuracy
of energy budgets (Fig. 3). Thus the gas dilution method may have been the only nondestructive, body composition method accurate enough for this study (Henen, 1991, 2001).
Lipid catabolism helped females survive winter dormancy but lipids remained unchanged during each
spring (Figs. 2, 3). Body lipid mass and energy were
higher at the end of both years compared to values at
the beginning of the study. These results, plus the increase in total body energy during the first year (Fig.
3), suggest that lipid and energy availability did not
limit egg production in either year (Henen, 1997).
Chelonians tend to have very little adipose tissue
(Pond, 1978). Desert tortoises appear to have little if
any adipose tissue, but vitellogenic follicles may contain much of the lipids in female G. agassizii (unpublished observations, B.T.H.), so lipids are important to
egg production. However, some chelonians may depend heavily upon body lipid reserves to complete vitellogenesis and egg production (Congdon and Tinkle,
1982; Long, 1985). Yolk lipids can also represent important maternal investments in neonatal survival
(Congdon and Tinkle, 1982), revealing the nutritional
and evolutionary complexity of life histories. Regardless, the availability of lipids or energy, relative to protein (nitrogen) and water availability, may rarely limit
egg production in desert tortoises (Henen, 1994, 1997).
Drinking in August 1987 enabled females to eat dry
grass and store lipids at rates more than 10-fold that
of pre-drinking rates (Fig. 3). However, converting dry
grass to body lipids was costly in terms of body water
and nonlipids (e.g., digestive mucosa and enzymes;
Fig. 2; Henen, 1994; Nagy et al., 1998). Using body
water and nonlipids to accrue lipids in summer may
be an inherent metabolic preparation for winter that
conserves body nonlipids during winter (Henen,
1997). Nonetheless, the large loss of nonlipid dry matter (probably protein) in summer 1987 probably reduced reproductive output in 1988. Other animals also
anabolize and catabolize protein in response to changes in dietary nitrogen availability (Henen, 1997). This
mechanism is consistent with G. agassizii’s propensity
to relax homeostasis (Nagy and Medica, 1986; Henen,
—
—
—
—
—
—
12.1
—
—
—
3.6
7.3
Rep
Non
21.6
23.5
3.0
17.5
18.4
268
257
33.8
18.4
268
173
375
—
—
—
—
—
—
0.12
—
—
—
0.04
0.07
Rep
Non
0.41
0.50
0.08
0.31
0.32
2.16
2.07
0.32
0.32
2.16
1.39
3.02
19.4
21.5
3.3
14.1
14.7
19.3
—
14.5
14.7
19.3
3.9
6.9
12.3
15.7
4.0
4.6
21.4
45.7
46.4
17.5
21.4
45.7
34.1
52.7
Non
Decrease
change %
Rep
Desert tortoise
Rate
% per day
0.20
0.23
0.03
0.14
0.15
0.21
—
0.15
0.15
0.21
0.04
0.07
Rep
0.14
0.18
0.04
0.05
0.17
0.47
0.47
0.18
0.17
0.47
0.35
0.54
Non
Rate
% per day
44.0
62.7
5.0
50.9
58.8
181
69.9
84.9
58.8
181
52.1
77.0
Rep
9.1
12.7
3.4
14.9
9.9
64.8
52.5
43.5
29.7
98.0
38.1
58.9
Non
Increase
change %
Rep
4.80
2.24
0.22
1.82
2.18
6.47
2.50
3.14
2.18
6.71
1.93
2.85
Non
0.33
0.45
0.12
0.53
0.36
1.85
1.50
1.61
0.96
3.63
1.41
2.12
44.0
41.9
3.3
51.5
40.9
76.2
58.0
52.4
40.9
76.2
50.1
59.7
Rep
12.0
11.9
1.9
11.4
8.4
21.2
10.7
29.3
25.3
52.0
40.2
47.0
Non
Decrease
change %
Size-blotched lizard
Rate
% per day
Rep
6.10
3.96
0.12
1.48
1.52
6.63
4.09
1.94
1.50
2.80
1.86
2.21
0.47
0.47
0.07
0.46
0.34
0.85
0.43
0.95
0.82
1.90
1.49
1.74
Average
Non
Rate
% per day
0.0
13.0
10.0
0.0
0.0
89.3
95.8
38.1
0.0
89.3
100.0
39.1
39.6
ANOVA
comparison %
a
* The final column indicates percent of significant RMANOVA multiple comparisons for G. agassizii that were detected via standard ANOVA multiple comparisons.c
Reproductive season—mid-April to mid-July; percent lipids increased due to large declines in nonlipids.
b Calculated from female lizard data in Figure 2.2, Table 2.1 and body energy values, of Nagy (1983). Reproductive data: late February to late June. December values were calculated
averages of November and January values. Percent body lipid equaled the sum of constituent lipids calculated from constituent masses and lipid contents (i.e., extractable somatic lipid,
100% lipid; reproductive tissues, 40.2% lipid; fat bodies, 74.4% lipid between May and September, 83.2% lipid otherwise). Lipid percentages for fat bodies and reproductive tissues
interpolated from tissue energy densities and measured energy densities of lipids (38.1 kJ/g) and nonlipids (17.7 kJ/g for a 50:50 carbohydrate-protein mix; Nagy, 1983).
c Figures 1–3 indicate the number of significant comparisons for the RMANOVA.
d L 5 lipid, NL 5 nonlipid, L/NL 5 percent change in ratio of lipid energy to NL energy; percent categories are value as percent of body mass (Body water, %; Body lipid, %) or
total body energy (Lipid energy, %).
Body mass
Body water
Body water, %
Dry mass
NL dry massd
Lipid mass
Body lipid, %a
Body energy
NL energy
Lipid energy
Lipid energy, %
L/NL, energy
Increase
change %
TABLE 1. Maximum increase and decrease (percent change and percent change/d) of mean composition values of female Gopherus agassiziia and Uta stansburianab during the reproductive
season (Rep) and the rest of the year (Non).*
46
BRIAN THOMAS HENEN
TORTOISE REPRODUCTIVE NUTRITION
FIG. 2. Mean and 95% CI for dry (total), nonlipid dry and lipid
mass, plus percent lipid (% of body mass), of female tortoises. These
parameters changed through time (RMANOVA F9,86 . 4, 7, 59, 51
and P , 1023, 1026, 10228, 10226, respectively). Sample size, dates
and significant comparisons indicated as in Figure 1.
1994; Peterson, 1996; Henen et al., 1998; Christopher
et al., 1999) and appears critical to desert tortoise survival and reproduction.
Metabolism of nonlipid dry matter
Nonlipid dry matter supported winter 1987–88 metabolism and was forfeited each spring (1988: 65 g;
1989: 51 g), with 25.2 and 24.5 g, respectively, being
invested in eggs. The remainder of these spring declines was probably realized through the turnover of
digestive enzymes and mucosa (Nagy et al., 1998) and
urate excretion (Nagy and Medica, 1986; Oftedal and
Allen, 1996; Nagy et al., 1998).
Nonlipid dry matter decreased 11% during the first
year (Figs. 2 and 3). Additionally, females laying eggs
in 1989 forfeited nonlipid dry matter (22 g or 0.21
MJ) during the second year while nonreproductive females increased nonlipid dry matter (74 g or 0.73 MJ;
Henen, 1994, 1997). Limited nitrogen consumption
before hibernation constrains egg production, apparently by limiting prehibernatory development of ovarian follicles (Henen and Oftedal, 1999). Thus, prehibernatory nitrogen consumption in 1982–1984 may
have helped the 1983–1985 egg production appear independent of rainfall and food availability. Biologists
have long suspected that protein (nitrogen) availability
limits avian egg production (Drent and Daan, 1980).
Homeostasis, reproductive effort and bet-hedging
Body energy fluxes (total) were substantial portions
of energy budgets. The fluxes represented up to 85%
of positive, and 400% of negative, rates of metabolizable energy intake. These fluxes also exceeded field
metabolic rates (FMR) in four intervals, and exceeded
reproductive allocations in spring of both years (Henen, 1997). The 10-fold seasonal differences in FMR
were reasonable for an ectothermic poikilotherm.
However, early spring FMR differed 10-fold between
47
FIG. 3. Mean and 95% CI for total (TOT), nonlipid (NL) and lipid
(L) energy content, plus % of total as lipids, of female tortoises.
These parameters changed through time (RMANOVA F9,86 . 17, 7,
59, 61 and P , 10214, 1026, 10228, 10228, respectively). Sample size,
dates and significant comparisons indicated as in Figure 1.
years (1989 , 1988; Henen, 1997) as females restricted surface activity to avoid the drought and poor forage conditions in 1989. This helped some females conserve enough nutrients to produce eggs (Henen, 1997).
Females rarely achieved energy balance, which they
did in spring of both years, illustrating their relaxed
homeostasis towards energy.
Reproductive effort (RE) is the proportion of energy
available to an organism that it allocates to reproduction (see Congdon et al., 1982; Nagy, 1983). For this
study, available energy equaled annual metabolizable
energy intake (MEI). Reproductive allocations included the chemical potential energy in eggs (Rc) and the
portion of FMR, Rr, related to reproduction (e.g., nest
digging and egg synthesis; Henen, 1997). Reproductive effort (RE 5 100 3 [Rc 1 Rr]/MEI) for the second
year, the drought year, was twice that of the first year
(26 vs. 13%; Henen, 1997) which had moderate rainfall and food availability. Reproductive effort may increase with high food abundance (Swingland, 1977),
especially for homeothermic endotherms, due to increased reproductive allocations and relatively stable
metabolizable energy intake. However, the opposite
occurred for female G. agassizii, metabolizable energy
intake declined 60% in the second year while statistically, reproductive allocations were unchanged between years. Fecundity in desert tortoises peaks when
food and water are abundant (e.g., El niño conditions,
Henen, 1997; Henen et al., 1998; Wallis et al., 1999).
Reproductive effort may then also peak, revealing a
complex relationship between reproductive effort and
resource abundance.
Why did most females (n 5 6), during an extreme
drought, forfeit body water (15%) and nonlipid dry
matter (5%; Henen, 1994) to produce just a few eggs?
Deserts are unpredictable (Louw and Seely, 1982), so
females can not predict whether their eggs will hatch
48
BRIAN THOMAS HENEN
under favorable conditions. Females forgoing reproduction (n 5 3) were guaranteed to have zero eggs
hatch regardless of ensuing conditions. However, females producing at least a few eggs had a greater
chance than nonreproductive females, of having eggs
hatch into favorable conditions. This reproductive pattern is consistent with a bet-hedging life history in an
unpredictable environment (see Congdon et al., 1982;
Philippi and Seger, 1989; Stearns, 1992). Accordingly,
spreading small reproductive investments over several
rounds, essentially reducing the variance in reproductive allocations among bouts (years or clutches), can
increase average winnings (reproductive success or fitness).
Apparently, three females could not afford to forfeit
body condition to produce eggs in 1989, so they forfeited one round of reproductive wagering. Reproductive females had relatively large protein and water reserves and could hedge their bets, producing just a few
eggs, with a chance their hatchlings would emerge into
favorable conditions. This may explain why Mojave
Desert females tend to produce at least a few eggs
every year (Henen, 1997; Wallis et al., 1999). However, patterns of rainfall and tortoise reproduction are
different in the Sonoran Desert (Murray et al., 1996;
Henen et al., 2001).
Maximum changes
The highest percent changes in dry matter and nonlipid dry matter approximated those for body water
(Table 1), the most labile component on an absolute
scale (Fig. 1). However, the greatest relative changes
occurred in lipids. The increase in percent lipids during
the reproductive season is misleading, resulting from
body water and nonlipid dry mass decreasing more
rapidly than did lipid mass. Also, maximum body
mass, water and dry matter decreases occurred during
the reproductive season while most energy decreases
were greatest in the nonreproductive season. These
patterns emphasize the importance of 1) body water
and nonlipid dry matter, more than lipids, to reproduction, and 2) relaxed homeostasis to desert tortoise
survival, relative to reproduction. We might expect this
for a long-lived, iteroparous species where each reproductive bout is small compared to lifetime reproductive output. Young tortoises appear more vulnerable
than adults which have relatively low mortality rates
(ca. 1 to 21% per year; Turner et al., 1984b; Peterson,
1994; Nagy et al., 1997). Nonetheless, the natural success of G. agassizii in deserts can not overcome the
multitude of anthropogenic forces threatening the species (Fish and Wildlife Service, 1994).
Gopherus agassizii and Uta stansburiana: Contrasts
We can compare the body composition of female
desert tortoises (ca. 1 to 3 kg body mass) and wild,
insectivorous, female Uta stansburiana (side-blotched
lizards, 2 to 5 g, Nagy, 1983), both species were studied in the Mojave Desert. However, the gas dilution
method did not discriminate the tissue location of tor-
toise lipids, so comparisons required several calculations using the lizard data (Table 1).
The components of lizard body composition
changed more during the reproductive season than in
the nonreproductive season (Table 1). This was probably linked to the lizards’ small size, short lives (ca.
2 yr) and few reproductive bouts (Turner et al., 1984a;
Nagy, 1983) relative to G. agassizii (Turner et al.,
1986; Henen, 1997; Wallis et al., 1999). For the reproductive season, and each clutch within the reproductive season, body composition changes for desert
tortoises were small compared to those for the lizards
(this analysis). Conversely, increases in the nonreproductive season were typically lower for lizards than
for tortoises (Table 1), probably due to the importance
of body nutrient reserves to tortoise survival in the
nonreproductive season. Also, on an absolute scale, the
small lizards have low food requirements and may face
food shortages less frequently (see Nagy, 1983) than
other insectivores or desert tortoises.
Despite the extreme water lability of desert tortoises, their maximum percent change in body water did
not always exceed those of U. stansburiana. The large
changes seen in U. stansburiana were associated mostly with reproduction (see also Hahn and Tinkle, 1965),
whereas G. agassizii changes can be related to reproduction (Henen, 1994), digestion (Henen, 1994, 1997;
Nagy et al., 1998), or dehydration (Nagy and Medica,
1986; Henen, 1994; Peterson, 1996).
Body size may underlie most of these species differences. The lizards are small, warm quickly, become
active in winter, and subsist on small insects or, if necessary, annual plants (Nagy, 1983). The relatively
large mass and thermal inertia of female desert tortoises usually prevents winter activity but facilitates
their relaxed homeostasis. Chelonian reproductive patterns seem affected by body size (Swingland, 1977;
Congdon et al., 1982; Henen, 1997), with small species having short maturation periods and lives, few
reproductive bouts, and high reproductive effort.
Early hatching, rapid maturation and early production of clutches are highly favored in U. stansburiana
(Hahn and Tinkle, 1965), selecting for early, rapid and
extensive conversion of fat body lipids into yolk material (Hahn and Tinkle, 1965; Nagy, 1983). This explains much of the large body composition changes in
the reproductive season and high reproductive effort
(45 to 84%; Nagy, 1983) of female U. stansburiana.
In contrast, female desert tortoises may reproduce
nearly every year for 20 to 50 yr. Although U. stansburiana rely heavily upon body reserves to produce
their first clutch of eggs, clutch frequency depends on
the current year’s food abundance (Turner et al.,
1984a).
Capital and income breeders
The capital, income and mixed reproductive strategies, originally described for birds (Drent and Daan,
1980), have been applied to other organisms including
chelonians (Kuchling, 1999). Biologists may consider
TORTOISE REPRODUCTIVE NUTRITION
desert tortoises ‘‘capital’’ breeders because female tortoises 1) forgo egg production and reabsorb follicles
if food availability is poor and body reserves are small
(Henen, 1993, 1994, 1997; Rostal et al., 1994), 2) accumulate water and protein reserves prior to winter
and use these reserves to help produce eggs (Henen,
1994, 1997; Henen and Oftedal, 1999), 3) have fullsized (‘‘preovulatory’’) follicles before entering hibernation (Rostal et al., 1994; Henen and Oftedal, 1999),
and 4) can emerge from winter dormancy and ovulate
prior to eating (Henen and Oftedal, 1999). However,
females can also produce more than one clutch, and
larger clutches, if 1) spring forage is abundant (Turner
et al., 1986, 1987; Henen, 1993; Wallis et al., 1999)
and 2) females lay their first clutch early in the reproductive season (Wallis et al., 1999). This suggests that
food (or nutrient) income also influences egg production, and that desert tortoises use a mixed reproductive
strategy. Yet, further research is necessary to determine whether females must accrue enough ‘‘capital’’
to meet new or different thresholds for producing additional clutches or larger clutches. Similarly, only further research will quantify the body condition threshold necessary to initiate vitellogenesis (see Kuchling,
1999).
Value of accurate repeated measures
The three nondestructive field methods were critical
for 1) ecologically relevant results (Nagy, 1980, 1983;
Henen, 1991, 1994, 1997), 2) obviating massive chemical extractions on numerous carcasses (ca. 90 females
5 9 females per sample 3 10 samples), 3) conserving
members of this Threatened Species, and 4) enabling
the use of repeated measures statistics. I also analyzed
the desert tortoise data by standard one-way ANOVA
(all df 5 9, 86) followed by Student-Newman-Keuls
multiple comparisons. If the data were not normal, I
used a one-way analysis of variance on ranks (df 5
9), followed by Dunn’s multiple comparisons. The repeated measures ANOVA was much more powerful
than standard ANOVA in detecting body composition
changes through time (Table 1).
The chosen methods were essential for assessing
physiological (e.g., relaxed homeostasis, nutrition and
body composition), ecological (e.g., reproductive effort, resource abundance, principle of allocation) and
evolutionary (e.g., life history) features of female desert tortoises. The results of Henen (1994, 1997) have
stimulated the use of other cutting-edge technologies
to better understand the dietary constraints upon the
reproductive cycles of female desert tortoises. These
technologies include ultrasonography (Henen and Oftedal, 1999; Henen et al., 2001; see also Rostal et al.,
1994 and review by Kuchling, 1999) and stable isotope analyses (unpublished data, B. T. Henen and O.
T. Oftedal; see Godley et al., 1998). This illustrates
how science and technology advance synergistically in
an alternating, iterative fashion.
49
ACKNOWLEDGMENTS
I am extremely grateful for help from G. A. Bartholomew, K. H. Berry, M. D. Hofmeyr, K. A. Nagy,
O. T. Oftedal, V. H. Shoemaker, F. B. Turner, U2, H.
van Wyk and especially D. L. Yee-Henen. The California Department of Fish and Game kindly allowed
me to complete these field studies (Memoranda of Understanding). I also thank D. Goldstein and B. Pinshow
for their support and the Society of Integrative and
Comparative Biologists and National Science Foundation for sponsoring this symposium.
REFERENCES
Christopher, M. M., K. H. Berry, I. R. Wallis, K. A. Nagy, B. T.
Henen, and C. C. Peterson. 1999. Reference intervals and physiologic alterations in hematologic and biochemical values of
free-ranging desert tortoises in the Mojave Desert. J. Wildlife
Dis. 35(2):212–238.
Congdon, J. D. and D. W. Tinkle. 1982. Reproductive energetics of
the painted turtle (Chrysemys picta). Herpetologica 38(1):228–
237.
Congdon, J. D., A. E. Dunham, and D. W. Tinkle. 1982. Energy
budgets and life histories of reptiles. In C. Gans and F. H. Pough
(eds.), Biology of the Reptilia, Vol. 13, Chapt. 5, Physiology D,
pp. 233–271. Academic Press, New York.
Drent, R. H. and S. Daan. 1980. The prudent parent: Energetic adjustments in avian breeding. Ardea 68:225–252.
Fish and Wildlife Service. 1994. Desert tortoise (Mojave population)
recovery plan. U.S. Fish and Wildlife Service, Portland,
Oregon, USA. 73 pp.
Gessaman, J. A., R. D. Nagle, and J. D. Congdon. 1998. Evaluation
of the cyclopropane absorption method of measuring avian body
fat. The Auk 115(1):175–187.
Godley, B. J., D. R. Thompson, S. Waldron, and R. W. Furness.
1998. The trophic status of marine turtles as determined by
stable isotope analysis. Mar. Ecol. Prog. Ser. 166:277–284.
Hahn, W. E. and D. W. Tinkle. 1965. Fat body cycling and experimental evidence for its adaptive significance to ovarian follicle
development in the lizard, Uta stansburiana. J. Exper. Zool.
158:79–85.
Henen, B. T. 1991. Measuring the lipid content of live animals using
cyclopropane gas. Am. J. Physiol. 261(30):R752–R759.
Henen, B. T. 1997. Seasonal and annual energy budgets of female
desert tortoises (Gopherus agassizii). Ecology 78(1):283–296.
Henen, B. T. 2001. Gas dilution methods: Elimination and absorption of lipid soluble gases. In J. R. Speakman (ed.), Body composition analysis: A handbook of non-destructive methods,
Chap. 4, pp. 99–126. Cambridge University Press, Cambridge,
England.
Henen, B. T. 1993. Desert tortoise diet and dietary deficiencies limiting tortoise egg production at Goffs, California [abstract]. In
Proceedings of the Desert Tortoise Council Symposium; 1992
February, p 97. Desert Tortoise Council, Las Vegas, Nevada.
Henen, B. T. 1994. Seasonal and annual energy and water budgets
of female desert tortoises (Xerobates agassizii) at Goffs, California. Ph.D. Diss., Univ. of California, Los Angeles. Available
from: University Microfilms, Ann Arbor, Michigan.
Henen, B. T., R. C. Averill-Murray, and C. M. Klug. 2001. Egg
follicles and yolks of Sonoran Desert tortoises (Gopherus agassizii)[abstract]. In Proceedings of the Desert Tortoise Council
Symposium, 2000 April, Desert Tortoise Council, Las Vegas,
Nevada. (In press).
Henen, B. T. and O. T. Oftedal. 1999. The importance of dietary
nitrogen to the reproductive output of female desert tortoises
(Gopherus agassizii)[abstract]. In Proceedings of the Second
Comparative Nutrition Society Symposium, 1998 August, pp.
83–88. Comparative Nutrition Society, Banff, Alberta, Canada.
Henen, B. T., C. C. Peterson, I. R. Wallis, K. H. Berry, and K. A.
Nagy. 1998. Desert tortoise field metabolic rates and water flux-
50
BRIAN THOMAS HENEN
es track local and global climatic conditions. Oecologia 117:
365–373.
Kenagy, G. J. and G. A. Bartholomew. 1985. Seasonal reproductive
patterns in five coexisting California desert rodent species. Ecol.
Monogr. 55(4):371–397.
Kuchling, G. 1999. The reproductive biology of the chelonia.
Springer-Verlag Berlin Heidelberg, Germany.
Lesser, G. T., W. Perle, and J. M. Steele. 1952. Measurement of total
body fat in living rats by absorption of cyclopropane. Am. J.
Physiol. 169:545–553.
Long, D. R. 1985. Lipid utilization during reproduction in female
Kinosternon flavescens. Herpetologica 41(1):58–65.
Louw, G. N. and M. K. Seely. 1982. Ecology of desert organisms.
Longman Group Limited, New York, New York.
Medica, P. A., R. B. Bury, and R. A. Luckenbach. 1980. Drinking
and construction of water catchments by the desert tortoise, Gopherus agassizii, in the Mojave Desert. Herpetologica 36(4):
301–304.
Minnich, J. E. 1982. The use of water. In C. Gans and F. H. Pough
(eds.), Biology of the reptilia, pp. 325–295. Academic Press,
New York.
Murray, R. C., C. R. Schwalbe, S. J. Bailey, S. P. Cuneo, and S. D.
Hunt. 1996. Reproduction in a population of the desert tortoise,
Gopherus agassizii, in the Sonoran Desert. Herpetol. Nat. Hist.
4:83–88.
Nagy, K. A. 1980. CO2 production in animals: Analysis of potential
errors in the doubly labeled water method. Am. J. Physiol. 238
(Reg. Integr. Comp. Physiol.) 7:R466–R473.
Nagy, K. A. 1983. Ecological energetics. In R. B. Huey, E. R. Pianka, and T. W. Schoener (eds.), Lizard ecology: Studies of a
model organism, pp. 25–54. Harvard University Press, Cambridge, Massachusetts.
Nagy, K. A. and D. P. Costa. 1980. Water flux in animals: Analysis
of potential errors in the tritiated water method. Amer. J. Physiol. 238 (Reg. Integr. Comp. Physiol.) 7:R454–R465.
Nagy, K. A., D. J. Morafka, and R. A. Yates. 1997. Young desert
tortoise survival: Energy, water and food requirements in the
field. Chelon. Conser. Biol. 2(3):396–404.
Nagy, K. A., B. T. Henen, and D. B. Vyas. 1998. Nutritional quality
of native and introduced food plants of wild desert tortoises. J.
Herpetol. 32(2):260–267.
Nagy, K. A. and P. A. Medica. 1986. Physiological ecology of desert
tortoises in southern Nevada. Herpetologica 42:73–92.
Oftedal, O. T. and M. E. Allen. 1996. Nutrition as a major facet of
reptile conservation. Zoo Biol. 15:491–497.
Peterson, C. C. 1994. Different rates and causes of high mortality
in two populations of the threatened desert tortoise Gopherus
agassizii. Biol. Conser. 70:101–108.
Peterson, C. C. 1996. Anhomeostasis: Seasonal water and solute
relations in two populations of the desert tortoise (Gopherus
agassizii) during chronic drought. Physiol. Zool. 69:1324–1358.
Philippi, T. and J. Seger. 1989. Hedging one’s evolutionary bets,
revisited. Trends Ecol. Evol. 4(2):41–44.
Pianka, E. R. 1978. Evolutionary ecology, 2nd ed. Harper and Row,
Publishers, New York.
Pond, C. M. 1978. Morphological aspects and the ecological and
mechanical consequences of fat deposition in wild vertebrates.
Ann. Rev. Ecol. System. 9:519–570.
Pough, F. H., C. M. Janis, and J. B. Heiser. 1998. Vertebrate life,
5th ed. Prentice Hall, Upper Saddle River, New Jersey.
Rostal, D. C., V. A. Lance, J. S. Grumbles, and A. C. Alberts. 1994.
Seasonal reproductive cycle of the desert tortoise (Gopherus
agassizii) in the eastern Mojave Desert. Herpetol. Monogr. 8:
72–82.
Schmidt-Nielsen, K. 1997. Animal physiology: Adaptation and environment. Cambridge University Press, Cambridge, UK.
Stearns, S. C. 1992. The evolution of life histories. Oxford University Press, Oxford, UK.
Swingland, I. R. 1977. Reproductive effort and life history strategy
of the Aldabran giant tortoise. Nature 269:402–404.
Turner, F. B., P. A. Medica, K. W. Bridges, and R. I. Jenrich. 1984a.
A population model of the lizard Uta stansburiana in Southern
Nevada. Ecol. Monogr. 52(3):243–259.
Turner, F. B., P. A. Medica, and C. L. Lyons. 1984b. Reproduction
and survival of the desert tortoise (Scaptochelys agassizii) in
Ivanpah Valley, California. Copeia 1984:811–820.
Turner, F. B., P. Hayden, B. L. Burge, and J. B. Roberson. 1986.
Egg production by the desert tortoise (Gopherus agassizii) in
California. Herpetologica 42(1):93–104.
Turner, F. B., K. H. Berry, D. C. Randall, and G. C. White (Environmental Biology, University of California, Los Angeles).
1987. Population ecology of the desert tortoise at Goffs, California, 1983–1986. Report to the Southern California Edison
Company, Research and Development Series # 87-RD-81.
Wallis, I. R., B. T. Henen, and K. A. Nagy. 1999. Egg size and
annual egg production by female desert tortoises (Gopherus
agassizii): The importance of food abundance, body size and
date of egg shelling. J. Herpetol. 33(3):394–408.