Ecology (1976) 57: pp. 445-458
ECOLOGICAL AND PHYSIOLOGICAL ASPECTS OF
REPRODUCTIVE STRATEGIES IN TWO LIZARDS'
W. KENNETH DERICKSON2
Division of Biology, Kansas State University, Manhattan, Kansas 66506 USA
Abstract. Two lizard species, the northern prairie lizard (Sceloporus undulatus garmani) from Reno County, Kansas and the northern sagebrush lizard (Sceloporus graciosus graciosus) from Washington County, Utah, were used to test four hypotheses and one assumption related to the theory of r- and K-selection. The northern prairie lizard is short-lived, matures early, and has a high reproductive effort (r-strategist) while the northern sagebrush lizard is long- lived, has delayed maturity, and a low reproductive effort (K-strategist). One of the assump- tions of the r- and K-selection theory is that competition for food is more intense for K- strategists than for r-strategists. Given this assumption, greater food availability for the prairie lizard was hypothesized to result in (1) a higher level of body lipids, (2) a higher rate of lipid utilization, (3) a lower percentage of ingested energy available for metabolism, and (4) an expenditure of less energy per offspring than in the sagebrush lizard.
Total lipid levels in the two species collected before and after hibernation indicated that prairie lizards had significantly higher lipid levels than sagebrush lizards during both collection periods. A comparison of the amount of body lipids lost and the amount of egg lipids gained during vitellogenesis of the first clutch suggested that more lipids are being utilized for egg production in prairie lizards than in sagebrush lizards.
Sagebrush lizards apparently are better adapted physiologically to lower food levels since they had lower rates of lipid utilization during starvation studies and they extracted more energy usable for metabolism from ingested food than prairie lizards.
Although both species expended about the same amount of energy on a given clutch of eggs, prairie lizards produced more offspring per clutch and therefore expended less energy per offspring. Prairie lizards produced as many as three clutches per season as compared to two clutches per season for sagebrush lizards, and changes in length-lean weight relationships for the two species over a given season suggested that the additional clutch of eggs produced by prairie lizards may be contributing to the higher mortality observed in this species.
Indirect evidence, based on precipitation levels and insect biomass at the two study sites and mouthgapes of the two species, supported the assumption that more food was available to the prairie lizards than the sagebrush lizard. Higher precipitation levels (indicating greater insect biomass) occurred at the prairie lizard collection site and this species had a smaller mouthgape index, indicating greater specialization in utilization of prey sites.
Key words: Energy, available for metabolism, expended per offspring; Kansas; lipids; prairie swift; sagebrusli lizard; strategy, reproductive; Utah.
INTRODUCTION
Although a number of studies have been done on
the importance of lipids in the life histories of
various animals, no one has attempted to correlate
differences in lipid storage and utilization with life
history differences. In most seasonal animal spe-
cies, lipids should be important since lipid storage
is a biochemically efficient way to package energy
for later usage. A gram of stored lipids represents
38 kJ (9 kcal) of energy, while a gram of carbo-
hydrate or protein, with the higher H,O content,
represents only 17-21 kJ (4-5 kcal) of energy.
Therefore, more energy can be stored in a smaller
package with lipids. In nonseasonal (acyclic) animal
'Manuscript received 10 June 1974; revised 19 De- cember 1975; accepted 19 January 1976.
2 Present address: Division of Environmental Impact Studies, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439 USA.
species, lipid storage is probably not important since
food is always presumed to be limiting but never
critically short (e.g., reproduction still occurs at low levels).
Nikolskii (1969) has shown that a correlation
exists between lipid levels and fecundity in a num- ber of commercial fishes. Odum (1960), Helms (1968) and King (1970) demonstrated marked in- creases in lipid storage prior to migration in several passerine bird species. Jameson and Mead (1964) and Davis (1967) indicated that lipids are used during hibernation in several small mammal species.
Sawicka-Kapusta (1968) found lipid depletion asso-
ciated with reproduction in field mice. A vast body
of literature on lizards indicates that some species use
lipids during hibernation (Dessauer 1955; Mueller
1969, Derickson 1974), while others use lipids during
reproduction (Hahn and Tinkle 1965, Licht and Gor-
man 1970, Telford 1970). Church (1962) could find
no evidence of lipid deposition and utilization in
446 W. KENNETH DERICKSON Ecology, Vol. 57, No. 3
three species of Jamaican House Geckos which live in a nonseasonal environment.
The literature on life history phenomenon is also bountiful. Several authors (Cole 1954; MacArthur and Wilson 1967; Gadgil and Bossert 1970; Hairston
et al. 1970; Pianka 1970, 1972) have put forth arguments to explain variations in life history phe-
nomenon. As a result of work done by these authors it is known that a gradient of life history strategies (strategy being used in a teleological sense) exists,
with the two extreme strategies being organisms with short lives, high reproductive efforts, and early ma-
turity and those with long lives, low reproductive effort, and delayed maturity. Tinkle (1969), Tinkle et al. (1970), and Gadgil and Solbrig (1972) have presented empirical evidence on lizards and dande- lions, respectively, that supports this dichotomy of life history strategies. This dichotomy has been more
popularly referred to as r- and K-selection (Mac- Arthur and Wilson 1967; Gadgil and Bossert 1970; Pianka 1970, 1972; Gadgil and Solbrig 1972), al-
though Hairston et al. (1970) choose to call it b and d selection. Basically, the theory of r- and K-selec- tion states that fitness is determined by r (intrinsic rate of increase) in species living in an environment where mortality is unpredictable, while in species living in environments where mortality is more pre-
dictable, fitness is determined by K (environmental carrying capacity). In other words, fitness is deter- mined primarily by density independent factors in r-
selected species and by density-dependent factors in K-selected species. For lack of better terminology, r- and K-selection will be used; however, the inde-
pendent variable identified as the causative agent is food availability per individual. This includes not only the potential productivity available but also an organism's access to that productivity. This avoids the issue of whether r-selected species are regu- lated by density-independent factors and K-selected species by density-dependent factors. Also, to avoid another problem with the theory it is assumed that the adult and juveniles of the species studied are both utilizing similar resources. Other assumptions are (1) that an optimum size for reproduction exists and that greater food availability will result in reach- ing this size at an earlier age; (2) that there is a positive correlation between food availability and reproductive effort; and (3) that an inverse rela- tionship between reproductive effort per season and life span exists.
Since differences in life history strategies and in
patterns of lipid storage and utilization are known to
exist in lizards, it seems logical to conclude that
the two may be interrelated. In other words, given
the assumption that differences in food availability
per individual can result in a variety of life history
strategies, it is likely that this same factor may also
explain observed differences in lipid storage and
utilization.
Differences in food availability per individual
should also lead to differences in (1) rates at which
lipids are utilized during periods of food shortage,
(2) the percentage of ingested energy available for metabolism, and (3) the amount of energy expended
on each offspring. An individual of a species that
has historically been exposed to low food availabilities should maximize its usage of available food. This may
be done by maximizing net energy yield per unit of feeding time, as suggested by Emlen (1966), Mac- Arthur and Pianka (1966), MacArthur and Wilson
(1967) and Schoener (1971), through utilization of prey with high caloric values or by utilizing prey that can be captured with a minimum of pursuit and cap- ture time. Given no difference in quality or captur-
ability of prey, a species that has been exposed to lower food levels may maximize energy obtained from
ingested energy by (1) reducing losses of ingested
energy via respiration, secretions, and excretions, and (2) by conserving stored energy during periods of food shortage. On the other hand, individuals of a species that has historically been exposed to abundant food supplies need to be less stringent in their conservation of stored energy during periods of food shortage and in their utilization of ingested
energy.
Food availability to the offspring may dictate how a parent apportions the ingested and stored energy available for reproduction. If the offspring and adults of a given species have both been exposed to low levels of food, a parent may increase its fit- ness by expending more energy per offspring in the form of parental care or by producing larger off- spring (Smith and Fretwell 1974). This assumes
that though fewer offspring are produced, their
survivorship is enhanced by an increased competitive
ability. If, on the other hand, food is abundant to
both adults and offspring, the parent may increase
its fitness by expending less energy per offspring.
This would result in the production of more offspring
and it is assumed that survivorship is enhanced by
sheer numbers. Harper et al. (1970) indicated that
plants under intense competition put more energy
into each offspring and produce fewer of them. Dr.
Eric R. Pianka (personal communication) compiled
a list of various lizard species and indicated that there
tends to be a negative relationship between the
amount of energy expended per offspring and the
number of eggs produced.
By using a r- and K-selected lizard species, an at-
tempt will be made to verify the following hy-
potheses:
Late Spring 1976 ASPECTS OF REPRODUCTIVE STRATEGIES IN LIZARDS 447
1) Given greater food availability for the r-se-
lected lizard species, this species should have
higher lipid levels before and after hibernation
than the K-selected lizard species.
2) Given greater food availability for the r-se-
lected lizard species, this species should have
a higher rate of lipid utilization during periods
of food shortage than the K-selected lizard
species.
3) Given greater food availability for the r-selected
lizard species, this species should obtain less
energy for metabolism per unit of ingested
food than the K-selected lizard species.
4) Given greater food availability for the r-se-
lected lizard species, this species should pro-
duce more offspring with less energy expended
per offspring, while the K-selected lizard spe-
cies should produce fewer offspring with more
energy expended per offspring.
SPECIES STUDIED
Females of the lizards Sceloporus undulatus gar- inani and Sceloporus graciosus graciosus were chosen
for this study. Some of the reasons for choosing
lizards and these particular species were (1) they
are easy to study either in the field or laboratory; (2) information related to the theory of the evolution of life history strategies in lizards is available from
work done by Tinkle (1969) and Tinkle et al. (1970); (3) a number of studies have looked at the
role of lipids in lizards; (4) reproductive effort can
be better approximated (since lizards exhibit no pa- rental care, kilojoules per clutch and number of clutches per season should provide a reasonably ac- curate measure of reproductive effort); (5) these species are similarly sized; thus, effects of body size differences on metabolism are minimized; and (6) in terms of lifespan, age at maturity, and reproductive effort, these two subspecies are known (Ferguson and Bohlen 1973, on S. u. garmani and Tinkle 1973, on S. g. graciosus) to fit the r- and K-selection corre- lates of Pianka (1970).
Sceloporus undulatus garmani is a small iguanid
lizard found predominantly in eastern Colorado, Kansas, Nebraska, and Oklahoma. This subspecies, hereafter referred to as the northern prairie lizard, is found in a variety of habitats, including sandy areas, sandstone cliffs, sparse grass, woodpiles, and old trash piles (Smith 1946). At the collection site, located in Reno County, Kansas, they were found in grazed sand prairie around old fallen cottonwood (Populus sp.) trees and refuse piles.
Prairie lizards feed on arachnids, Coleoptera, Hy-
menoptera, Diptera, Orthoptera, and Hemiptera (Smith 1946). Their feeding strategy is to sit and
LaJ
m SCELOPORUS GRACIOSUS GRAC/OSUS
z2 I H S '1-I m C H s
SCEZOPORUS UNDULATUS GARMANI
co H S D
ti H S D
J A SON D J F M A MJJ A NDJFMAMJJASONDJ MONTH
FIG. 1. Time required to reach sexual maturity and the number of clutches produced seasonally by adults originating from the initial and final clutches of a season of Kansas northern prairie lizards, Sceloporus undulatus garmani and Utah northern sagebrush lizards, S. gracio- sus graciosus. Clear areas represent periods of either
maintenance, growth, or vitellogenesis. Stippled and cross-hatched areas represent periods of hibernation and reproduction, respectively. Hatching, sexual maturity, and death are indicated by the letters H, S, and D, re- spectively. Only 2 yr out of potentially 6 are shown graphically for S. g. graciosus. These graphs are partly based on data from studies done by Ferguson and Bohlen (1973) and Tinkle (1973).
wait until prey come within striking distance and
then give pursuit. The average lifespan is 1 yr, age
at maturity 8-9 mo, and reproductive effort is about
three clutches/season (Ferguson and Bohlen 1973). An early age at maturity and production of three
clutches results in considerable variation in egg-laying
times (Fig. 1). Offspring from the first clutch hatch
in mid-July, the second in mid-August, and the third
in early September. Lizards from the first clutch
will be ready to reproduce the following spring,
while those from succeeding clutches will not mature
until late spring or early summer. First clutch off-
spring have the potential for laying clutches in May,
June, and July of their first breeding season.
Probably, because of this high reproductive effort
at such a young age, the first clutch offspring rarely
reproduce a second season. Conversely, offspring
from the second and third clutches may produce 1-2
clutches in their first breeding season. These off-
spring from later clutches are more likely to survive
a second winter (but most do not) and produce addi-
tional clutches of eggs. Therefore, an early age at
maturity and the production of three clutches by
some lizards results in considerable variation of lay-
ing times and size of lizards laying eggs.
Sceloporus graciosus graciosus is found predomi-
nantly in Idaho, Nevada, Utah, and Wyoming at
altitudes > l,500 m. As the common name, northern
sagebrush lizard, suggests, this species is common
448 W. KENNETH DERICKSON Ecology, Vol. 57, No. 3
in sagebrush communities, but also may be found in open flat lands, bouldered regions, and forested slopes (Smith 1946). Tinkle (1973) studied this
subspecies in the Kolob region of Zion National Park in southwestern Utah. The Kolob includes
sandy sagebrush (Artemisia sp.) flats and sandy areas
between extensive areas of eroded sandstone. My collection site was a few kilometers east of Zion Na-
tional Park East Gate. The altitude is comparable to
that of Tinkle's study area, 2,000 m, and is char- acterized by sandy sagebrush flats.
Sagebrush lizards have the same general diet and foraging patterns as prairie lizards. They are known
to be long-lived, > 6 yr, and mature in 2 yr (Stebbins and Robinson 1946, Tinkle 1973). The number of
clutches produced probably varies with altitude and latitude, however, in southwest Utah this species lays
two clutches of about four eggs each season (Tinkle 1973). This reproductive schedule will allow an
average adult to produce about eight clutches in a lifetime. Unlike prairie lizards, sagebrush lizard hatchlings from first and second clutches reach re- productive maturity at the same time, the beginning of the third season. This results from an extra sum-
mer of growth prior to reproduction, which allows all hatchlings to reach reproductive size before their third season (Fig. 1) .
MATERIAL AND METHODS
Prairie lizards and sagebrush lizard females were collected over a 2-yr period (1972-1973) from Reno County, Kansas and Washington County, Utah, respectively. When possible 10 individuals of each species were collected after hibernation, during the vitellogenic and gravid periods of each clutch of eggs, and about 1 mo after reproduction had ceased. These collection periods were based on observed lipid cycling patterns in other lizard species, such as Anolis carolinensis (Dessauer 1955).
Field samples
Seventy-eight prairie lizards and 88 sagebrush liz- ards were collected in the field and kept cool until transferred to the laboratory, where they were quick- frozen immediately and analyzed at a later date for total lipids (simple and complex). The carcass (minus viscera) and eggs of each lizard were weighed and dried to a minimal and constant weight in a thermal vacuum oven at 40?C and 30 psi (Horwitz et al. 1970). Each part was then homogenized in 2: 1 vol/vol chloroform-methanol solution in a tissue homogenizer and the homogenates treated according to the Folch method (Folch et al. 1957). Weight of total lipids for each depot were recorded as a per-
centage of the lean (no lipids) dry body weight (LDBW) to correct for variations in body size. Pre-
liminary studies indicated that complex lipids, those
generally not available as storage lipids, represented < 10% of the total lipids. A similar figure for com-
plex lipids was found in Sceloporus jarrovi by Hadley and Christie (1974). Since complex lipids represent such a small percentage of the total lipids, no at-
tempt was made to exclude these lipids. Accordingly, all figures are slight overestimates of storage lipids
in both species.
Rates of lipid utilization were calculated for both
species from field samples by comparing lipid levels
in lizards that had come out of hibernation with those ready to lay their first clutch (gravid-initial
clutch). Individual lipid levels (mg/g LDBW) of the initial gravid sample was subtracted from the
mean lipid level of the posthibernation lizards and this figure was divided by the number of days be-
tween these two samples. The same method was used
to calculate rates of lipid deposition; only the mean
lipid level of the final gravid sample was subtracted from individual lipid levels in the postreproductive
sample.
Experimental studies
An additional 15 females of each species were collected in 1973 just after the lizards had finished laying their last clutch of eggs (mid-July for both species). Three lizards from each species were analyzed for total lipids to determine initial lipid levels. The remaining lizards of both species were placed in animal cages under similar moisture (watered every other day) and photoperiod regimes (14L: lOD) and were studied simultaneously to de- termine rates of lipid deposition when fed and utilization when starved. An infrared lamp was
placed in one corner of the cage to allow the lizards to thermoregulate. Each female was fed a surplus weight of crickets every other day. Feces, urates, and remaining crickets were removed and weighed before the next feeding. Feces and urates were frozen to prevent bacterial decomposition until they could be analyzed for energy content. When the lizards had gained 10% of their body weight, six were sacrificed and analyzed for total lipids as de- scribed above, to determine rates of lipid deposition for each species. Remaining lizards were starved until they had lost 10% of their body weight, then analyzed to determine rates of lipid utilization. By knowing the initial lipid levels (mg/g LDBW) and the number of days required to gain or lose 10%
of the body weight, rates of lipid deposition and
utilization could be calculated. The initial lipid
level for the fattening study was determined from the
initial sample of lizards mentioned above, while
the initial level for the starvation study was the
lipid level of the lizards that had gained 10% of their
Late Spring 1976 ASPECTS OF REPRODUCTIVE STRATEGIES IN LIZARDS 449
body weight. In both cases initial values were com- puted as means and the difference between these mean values and each starved or fattened lizard was determined. This calculation yielded the number of milligrams of lipid per gram LDBW gained or lost, which was then divided by the appropriate number of days to yield rates of deposition and utilization for the two species.
Calorimetry
Both feces and urates collected during the fattening study and crickets and oviducal eggs from the differ- ent clutches, were placed in a drying oven at 60?C and dried to a minimal and constant weight. Some fecal and urate pellets, crickets, and oviducal eggs were then placed in a muffle furnace at 500?C for 4 h to measure ash content. Remaining dried feces, urates, crickets, and oviducal eggs were analyzed for caloric value using a Parr Semi-micro Oxygen Bomb Calorimeter (Model 1411) ' according to techniques described in Paine (1971). All caloric values were
corrected for ash. From these data it was possible to calculate calories per clutch of eggs, calories per egg, ingested and egested energy. Calories per clutch was considered the major proportion of energy ex-
pended on reproduction since lizards exhibit no pa- rental care. The percentages of ingested energy available for metabolism (ME) and assimilation ef- ficiency (AE) were calculated using the formulae,
ME IE- (FE + NE) /IE X 100 and
AE IE- FE/IE X 100,
where IE is ingested calories, FE is fecal calories, and NE is nitrogenous waste calories. It was also possible to calculate what percentage of the ME was being stored as lipids since the ME and fattening studies were run concurrently. This was done by
converting the lipid increases in the fattening study to caloric values (a gram of lipid is 38 kJ, or 9 kcal), dividing this figure by the amount of ingested energy available for metabolism, and multiplying by 100.
Statistical analyses
Least squares analysis of variance was used to compare lipid levels, calories per clutch, calories per egg, eggs per clutch, ME, AE, and the proportion of the energy available for metabolism that was con- verted to lipids. Rates of lipid deposition and utili- zation represented deviations from the mean value of the initial samples and these deviations were also compared using least squares analysis of variance. Regression analyses were used to assess relationships between any of the above parameters and LDBW. If relationships existed regression lines were compared
both within and between species to determine sta- tistically significant differences. Chow's (1960) tech- nique was used to compare residual sums of squares of the individual regressions with that of the data combined into a single regression line. If weight was correlated with a given parameter in both species, this parameter was expressed on a per gram basis from the regression equations. If no relationship existed between LDBW and a given parameter in either species, the raw data were compared. The rationale for seeking a relationship between LDBW and each of the above parameters was based on observed relationship of clutch size and body weight in a number of lizard species.
Evidence for the assumption that food level differences exist between
prairie lizards and sagebrush lizards
In order to provide some support for this as- sumption the following comparisons were made: (1) precipitation levels in the two study areas over the past 30 years using U.S. Weather Bureau data; and (2) mouthgape (head length X mouth width) vs. snout-vent length for the two species. Insect biomass is correlated with precipitation (Janzen and Schoener 1968) and mouthgape with prey sizes utilized (Schoener 1968). At low food levels a spe- cies should use a wider array of prey sizes.
RESULTS
Hypothesis 1
This hypothesis stated that r-selected lizard species (prairie lizards) should have higher lipid levels prior to and after hibernation than K-selected lizard spe- cies (sagebrush lizards). My results support this hypothesis (Fig 2A). Total body lipids per gram LDBW, pooled over the two collection years, were higher prior to and after hibernation in prairie lizards than in sagebrush lizards.
In prairie lizards, minimal lipid levels reached dur- ing production of the middle clutch were lower than in sagebrush lizards. In sagebrush lizards, lipid levels were minimal during production of the final clutch. Neither species apparently depleted their storage lipids since minimal levels (6%-7% of LDBW) were higher than polar lipid levels (2% of LDBW). However, prairie lizards utilized more of their stored lipids during reproduction than sagebrush lizards.
Lipid levels changed more in prairie lizards than in sagebrush lizards during egg production. The loss of lipids from posthibernation to the initial gravid sample was approximately 200 mg/g LDBW in
prairie lizards versus about 60 mg/g LDBW in sage- brush lizards. Simultaneously, there was an increase
of about 210 and 220 mg of lipids in the eggs of
450 W. KENNETH DERICKSON Ecology, Vol. 57, No. 3
400
3003
-J'200 $ t1 4 LLJ
100-
B
400
9
300 100- 10 8 -.4 ~~~20 13
U,
0-.
-~200
0~~~2
m 20~~2
0 5+ 10
A
0 PH V 01 02 03 PR
SAMPLE
FIG. 2. Seasonal variation of mean lipid levels (per gram lean dry body weight, LDBW) in Kansas northern prairie lizards (open bars) and Utah northern sagebrush lizards (solid bars). Samples are posthibernation (PH), vitellogenesis (V), gravid-initial clutch (01), gravid- middle or initial clutch (02), gravid-final clutch (03). Horizontal bars represent the means; vertical lines the ranges; and vertical bars one standard error. Numbers above bars indicate sample sizes. (A) total body lipids; and (B) egg lipids.
prairie lizards and sagebrush lizards, respectively
(Fig. 2B). The relationship between body lipids
lost and egg lipids gained in both species is expanded
upon in Fig. 3. Body lipid loss was equivalent to egg
BODY LIPIDS LOST (mg/g LDBW) EGG LIPIDS GAINED(mg/g LDBW)
200 100 0 100 200
FIG. 3. Histogram comparing mean body lipid loss with mean egg lipid gain (mg/g lean dry body weight, LDBW) in field samples of Kansas prairie lizards (clear bar) and Utah sagebrush lizards (solid bar).
lipid gain in prairie lizards, while egg lipid gain far
surpassed body lipid loss in sagebrush lizards.
Presumably, sagebrush lizards had to rely more
on other sources of energy (i.e., ingested) to produce
egg lipids than did prairie lizards. In fact, some
prairie lizards had 30,125 J (7,200 cal) in stored
lipids which seems more than enough to produce an
initial clutch of eggs with a mean energy value of
22,175 J (5,300 cal). In sagebrush lizards maximum
lipid levels were equivalent to 15,062 J (3,600 cal),
less than a 21,338 J (5,100 cal) clutch of eggs.
Table 1 further emphasizes that sagebrush lizards
had to rely more heavily on external sources of
TABLE 1. Mean rates of body lipid loss (negative value) or gain and mean rate of egg lipid deposition in field and laboratory samples of Kansas prairie lizards (PL) and Utah sagebrush lizards (SL). LDBW - Lean Dry Body Weight; * = 0.05 level of significance; ** - 0.01 level of significance; NS - not significant
Body Eggs (mg/g LDBW day) (mg/g LDBW- day)
Period of lipid loss in field Posthibernation to gravid-
initial clutch
PL -4.5 4.0 ** **
SL -0.6 6.7
Period of lipid gain in the field Gravid-second clutch to
postreproduction PL 2.3 None
NS
SL 2.3 None
Laboratory samples
Starvation experiment PL -17.0 None
*
SL -6.5 None
Fattening experiment PL 3.2 None
NS
SL 3.6 None
Late Spring 1976 ASPECTS OF REPRODUCTIVE STRATEGIES IN LIZARDS 451
TABLE 2. Mean assimilation efficiency (AE), percent of ingested energy available for metabolism (ME), and percent of ingested energy available for metabolism stored as lipids (SF) in Kansas prairie lizards and Utah sagebrush lizards during fattening experiments. * - 0.05 level of significance
Prairie lizards Sagebrush lizards
AE n 12 86.16% n = 12 89.55% * *
ME n 12 76.17% n = 12 81.70% * *
SF n 4 12.51% n= 4 23.46% * *
energy than prairie lizards for egg production. The
rate of body loss (in milligrams per gram LDBW)
exceeded the rate of egg lipid deposition in prairie
lizards. In sagebrush lizards the rate of body lipid
loss was < 10% of the rate of egg lipid deposition.
Hypothesis 2
Based on this hypothesis, r-selected lizard species
(prairie lizards) should have higher rates of lipid
utilization than K-selected lizard species (sagebrush
lizards) during starvation. Data from the starvation
study supported this hypothesis (Table 1). The rate
of lipid utilization was almost three times greater in
prairie lizards than in sagebrush lizards. This dif-
ference suggests that sagebrush lizards were better
adapted to conserve lipids at lower food levels than
prairie lizards.
There was no significant difference between the
two species in the rate of lipid deposition in either
field or experimental samples, although rates of depo-
sition were higher in the laboratory than in the
field. This suggests that food is not unlimited for
either species in the field. Since both species have roughly the same amount of time to store lipids
prior to hibernation there is no reason to suspect
differences in rates of lipid deposition between the
two species.
Hypothesis 3
According to the third hypothesis, prairie lizards
should have a lower percentage of ingested energy available for metabolism than sagebrush lizards. Both
assimilation efficiency and percentage of ingested energy available for metabolism were higher in sage- brush lizards (Table 2). Not only were sagebrush lizards absorbing more of the ingested energy across the intestine, but a higher percentage of this assimi- lated energy was retained and not lost as nitrogenous waste (urates) . This higher assimilation efficiency and percentage of ingested energy available for metabolism suggests that sagebrush lizards are better adapted for lower food levels than prairie lizards.
TABLE 3. Mean calories/dry g of egg, eggs/clutch, calories/clutch, and calories/egg in Kansas prairie lizards (PL) and Utah sagebrush lizards (SL). Values for eggs/clutch and calories/clutch are functions of weight and were computed from linear regression equations. Caloric values were corrected for ash to eliminate eggshell thickness differences. NC = no clutch produced at this time; NS - not significant at the 0.05 level; * = significant at the 0.05 level; ** = significant at the 0.01 level. To convert calories to joules, multiply by 4.184
Clutch
May June July
Calories/dry g of egg
PL 6,088 6,453 6,657
SL NC 6,453 6,688
Eggs/clutch
PL 6.0 * 7.0 * 6.0 ** **
SL NC 4.4 * 3.1
Calories/clutch
PL 4,450 * 5,000 * 4,450 NS NS
SL NC 4,450 NS 4,450
Calories/egg
PL 677 * 734 * 854 ** **
SL NC 1,059 * 1,394
Hatchling size (mm)
PL 22.8 ** 24.1
SL NC
Prairie lizards not only were less efficient at utiliz-
ing energy, but a higher percentage of the energy
available for metabolism was devoted to metabolic
activities other than lipid storage. In sagebrush liz-
ards, a higher percentage of energy available for
metabolism was converted into lipids.
Since house crickets were utilized as an energy
source in this study, it is not known how accurately
these efficiencies compare to actual efficiencies.
However, the difference between the two species
should still be valid.
Hypothesis 4
This hypothesis stated that sagebrush lizards should
expend more energy per offspring than prairie lizards.
Regression equations of calories per clutch versus
LDBW were significant for both species (Fig. 4).
However, regression lines did not differ significantly
(Table 3). Therefore, lizards of both species put
about the same number of calories into eggs per
clutch. The regression of number of eggs per clutch versus LDBW also was significant for both species
(Fig. 5). Regression lines did differ significantly between species. A prairie lizard produced signifi- cantly more eggs per clutch than an equivalent-sized
sagebrush lizard (Table 3). Since the number
452 W. KENNETH DERICKSON Ecology, Vol. 57, No. 3
R:.992 .
iR=.877 11000 / !
/ / ,/-.8 R,'.989
9000 / /
7000 ~ ~ ~ ~ ~
/ .' I- /* /
3 7000
and /t hU-
5000-
0.0 1.0 .o2.0 3.0
LEAN DRY BODY WEIGHT (g)
FIG. 4. Calories/clutch versus lean dry body weight in Kansas prairie lizards (second clutch, s *; first and third clutch, ?--?) and Utah sagebrush lizards (first and second clutch, (--O)r . All three regres- sions were significant at the 0.01 level. The regression line for sagebrush lizards did not differ significantly, at the 0.05 level, from those for prairie lizards.
of calories per clutch was the same for both species
and the number of eggs per clutch was different for
the two species, calories per egg should and did
differ (Table 3).
In both species the calories per egg increased with
successive clutches. Thus more energy was expended
on offspring in later clutches. Egg quality (calories
per gram of eggs) also increased seasonally in both
species, but did not differ between species.
The number of calories per clutch per gram LDBW
and egg calories per body calories provide estimates
of the cost per clutch of reproduction to the parent
(Table 3). While the cost per clutch was about the
same for both species, prairie lizards produced three
clutches versus two for sagebrush lizards. Thus,
prairie lizards put considerably more energy into
eggs than sagebrush lizards each season (Table 4).
This additional expenditure of energy on egg produc-
tion by prairie lizards resulted in nonlipid tissue
loss (Fig. 6A). Comparing the regressions of LDBW
versus snout-vent length for posthibernation and gravid-final clutch samples in both species revealed
the prairie lizards lost more nonlipid tissue during re- production than sagebrush lizards (Fig. 6B). If there
is no net nonlipid catabolism, neither the slopes nor
=R.972
1 2
,j'R=.918
8~~~~~~~~~
m g<27 @ _@~~ ~ R =.839
I-. /~~~~~~~~A
0~~~~~~~~
0.0 1.0 2.0 3.0
LEAN DRY BODY WEIGHT (g)
FIG. 5. Eggs/clutch versus lean dry body weight in Kansas prairie lizards (second clutch, * ~*; first and third clutch, (D--(D) and Utah sagebrush lizards (first and second clutch, * - *). All regressions were significant at the 0.01 level. The regression line for sagebrush lizards differed significantly, 0.01 level, from those for prairie lizards.
the elevations of these lines should change through- out the season, since lean dry body weights were used. Therefore, any decrease in slope or elevation repre- sents nonlipid tissue loss. The greater nonlipid tis- sue loss in prairie lizards may render this species more vulnerable to predation or harsh climatic con- ditions, which would in part explain the shorter
TABLE 4. Estimated eggs/season, egg calories/season, eggs/lifetime, egg calories/lifetime, body calories, and egg calories /season-body calories in Kansas prairie lizards (PL) and Utah sagebrush lizards (SL). Eggs/ season and egg calories/season were calculated by summing the average eggs/clutch and calories/clutch, respectively, for all clutches. Eggs/lifetime and egg calories/lifetime were computed by multiplying eggs/ season and egg calories/season by the average number of breeding seasons that members of each species ex- perience. The values of 5,390 cal/gram and 5,300 cal/gram ash free dry body weight were used to cal- culate body calories for PL and SL, respectively. Figures in parentheses are on a per gram lean dry body weight basis. To convert calories to joules, multiply by 4.184
Parameters PL SL
Eggs/season 20.9 ( 19.0) 10.4 (7.5) Egg calories/season 15,457 (13,900) 12,945 (8,900) Eggs/lifetime 20.9 ( 19.0) 41.6 (30.0) Egg calories/lifetime 15,457 (13,900) 51,780 (35,600) Body calories 6,127 8,902 Egg calories/season- body calories 2.53 1.45
Late Spring 1976 ASPECTS OF REPRODUCTIVE STRATEGIES IN LIZARDS 453
/R=.992 3.0 /R.
_/ R=.995
I /
32.0 1"
C:1~~~~~~~~~~1
mU //
1.0-A ,
///
B
3.0
I-
2.0
> <= .9~~~~~~~86 m /
1.0- R _ =.985
40 5b 70 SNOUT-VENT LENGTH (mm)
FIG. 6. Lean dry body weight versus snout-vent length for posthibernation (0 0 ) and final clutch (*--*) in Kansas prairie lizards and Utah sage- brush lizards. (A) prairie lizards; (B) sagebrush lizards. All regressions were significant at the 0.01 level. Com- parisons between regression lines for both species were significant at the 0.01 level.
lifespan of females of this species. Since sagebrush
lizards are not expending as much energy on egg
production each season, this may reflect positively on
lifespan and enable sagebrush lizards to expend more
energy on egg production in a lifetime than prairie lizards (Table 4).
Assumption
The higher precipitation levels in Kansas than in Utah (Fig. 7) suggests that potentially more food
72
KANSAS
48
24
72
UTAH
48
24
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0+
INCHES OF PRECIPITATION
FIG. 7. Precipitation levels from 1944-1973 for Reno
County, Kansas and Washington County, Utah. The graphs represent the number of months out of the 30-yr
period that had a given level of precipitation.
was available to the r-selected lizard species (prairie
lizard) than the K-selected lizard species (sagebrush
lizard). Sagebrush lizards also had a larger mouth-
gape than equivalent-sized prairie lizards suggesting
that they were able to utilize a wider array of prey
sizes than prairie lizards (Fig. 8).
DISCUSSION
The results of this study suggest general models to
explain observed differences in patterns of lipid
storage and utilization of various lizard species.
Specifically, the r-selected lizard (prairie lizards)
stored more lipids than the K-selected lizard (sage-
brush lizards). Greater lipid stores resulted in more
lipids being available after hibernation in the r-se-
lected species, and these were apparently used in the
production of the first clutch of eggs. Because of the
lack of total dependence on ingested food to produce
the first clutch of eggs, the r-selected species is able to start reproducing earlier in the season which re-
sulted in a higher reproductive effort. This higher reproductive effort appeared to have a negative feed-
back on lifespan. Lower lipid levels in the K-selected
lizard species resulted in this species being more de-
pendent on ingested energy for egg production. Egg laying occurred later in the season and fewer clutches
were laid. This lower reproductive effort had less
of a negative feedback on life span than in the r-se-
lected species. The r-selected species also had a
higher rate of lipid utilization when starved, a lower
percentage of ingested energy available for metabo-
lism, and expended less energy per offspring than the
K-selected lizard species. To test the generality of
these findings, lipid storage and usage, rates of lipid
454 W. KENNETH DERICKSON Ecology, Vol. 57, No. 3
140 2x2
2 K
R R930
120- / * /
2~~~~~~~~ x/ / R.980 100~ ~ ~ ~~~~~
/20 02
80- / R 98 60
40 50 60 SNOUT-VENT LENGTH (mm)
FIG. 8. Mouthgape (head length x mouth width) versus snout-vent length in Kansas prairie lizards ( 0 * ) and Utah sagebrush lizards ( X- -X ). Both regressions were significant at the 0.01 level. Com- parisons between regression lines were also significantly different at the 0.01 level.
utilization, percentage of ingested energy available for metabolism, and energy expended per offspring in other lizards will be examined.
Supporting evidence from studies
on other lizard species
It is difficult to compare lipid usage data avail-
able on other lizard species with data in the present study. Most of the data available on lipid usage is anecdotal, and does not look at all potentially avail- able lipids, or at seasonal cycling of lipids. However, some data support my hypothesis that lizards with r-
strategist characteristics may have more lipids avail- able for reproduction. Based on limited data, Parker and Pianka (1973) observed no correlation between
reproductive activity and fatbody length/snout-vent length in female Sceloporus magister. They indicated that this species requires at least 2 yr to reach matu-
rity, produces 1-2 clutches/season, and is probably long-lived (K-strategist). Hahn and Tinkle (1965) and Telford (1970) found a significant decrease in fatbody lipids during reproduction in female Uta
stansburiana and Takydromus tachydromoides, re-
spectively. According to Tinkle (1969) and Tinkle et
al. (1970), these two species are short-lived, mature early, and have a high reproductive effort (r-strate- gists). Based on seasonal cycling of carcass, fatbody,
and liver lipids in female Cnemidophorus tigris,
Gaffney and Fitzpatrick (1973) estimated that >
3,000 calories of lipid go into egg production, as
compared to 1,500 calories for hibernation. These
authors collected their C. tigris from western Texas,
where they mature in 1 yr and produce two clutches/ season (r-strategist; Tinkle 1969). Female Ameiva
festiva and A. quadrilineata (Smith 1968) and
Cnemidophorus sexlineatus (Hoddenbach 1966) also showed decreases in fatbody size during reproduction.
These three species can be classified as r-strategists (Tinkle 1969, Tinkle et al. 1970). Licht and Gor- man (1970), Ruibal et al. (1972), and Gorman and
Licht (1975) found an inverse relationship between
reproductive activity and fatbody size in a number
of tropical anoles. Those species that reproduced
throughout the year either had no fatbodies or very
small ones, while those that exhibited reproductive
cycles had large fatbodies during the nonreproductive
phase (dry season) and lacked fatbodies while repro- ducing (wet season). Licht (1974) also found that
fatbodies in Anolis cristatellus would increase during
reproduction via supplemental feeding in the field.
Apparently, egg production and deposition in many
of these anoles creates too much of an energy de-
mand to allow lipid storage from available food. All
of the anoles studied by these authors could be classi-
fied as r-strategists.
Anolis carolinensis (an r-strategist; Tinkle 1969,
Tinkle et al. 1970) may be an exception to the
hypothesis that r-strategists use lipids primarily for reproduction. According to Dessauer (1955), this
species is active most of the year in New Orleans, but only reproduces from April to August. During the
months of November and December they are inactive
and lipid levels decreased markedly during this
period of time. Mean air temperature (and pre- sumably body temperature) in these 2 months is
relatively high (15?C) which may account for utili-
zation of large amounts of lipids. A higher hiberna- tion temperature than that of the other lizards
mentioned above would result in greater metabolic
demands and hence greater utilization of lipid. Des- sauer (1955) also stated that A. carolinensis feeds
little if at all during November and December. Lack
of feeding activity plus high winter body temperature
may impose a heavy drain on lipid reserves so that
lipid levels are too low to use for reproduction in the
spring.
Little or no data from the literature support my
results on rates of lipid utilization, percentage of in-
Late Spring 1976 ASPECTS OF REPRODUCTIVE STRATEGIES IN LIZARDS 455
gested energy available for metabolism (ME), and energy expended per offspring. This is the only study of rates of lipid utilization in lizards, so more empirical data on a number of species are needed to determine whether a low rate of lipid utilization is common to K-strategists. Some comparative data are available on assimilation efficiencies and ME, but it is difficult to draw conclusions from these. Mueller (1970) found a ME of 83% for the sagebrush lizard which is consistent with my results (82%). He also found a ME of 83% for Sceloporus occidentalis.
This lizard requires 2 yr to reach maturity; how- ever, it is intermediate to sagebrush lizards and prairie lizards regarding lifespan and clutch number (Tinkle 1969, Tinkle et al. 1970, Goldberg 1973). It is difficult to predict whether S. occidentalis should have a high or low ME. Mueller (1970) felt a ME of 83% was an overestimate, because S. occi- dentalis was studied much of the time at body tem- perature below the optimum for this species. Sage- brush lizards, however, were studied at their optimum temperatures. If there is an inverse relationship between temperature and ME, S. occidentalis should have a lower ME than sagebrush lizards. However, it is still difficult to categorize S. occidentalis as a r- or K-strategist.
Avery (1971) found that the lizard Lacerta vivip- ara, a viviparous K-strategist with delayed maturity and low reproductive effort (Tinkle et al. 1970), had an assimilation efficiency (AE) of 89%. In the present study, the sagebrush lizard had an AE of 90%. Finally, Johnson (1966) assumed an AE of 67% for S. undulatus, S. magister, and C. tigris when he measured the number of calories of energy as- similated per gram per day. However, these effi- ciencies are probably not accurate since my study shows variation in ME between species, and both my data and Pough's (1973) indicate that an AE of 67% is low for insectivorous lizards. Most in- sectivorous lizards have high efficiency percentages (from the low 70s to > 80), which may be due to the relatively high energy value of insects, 5,400 calories/ g (Golley 1961).
Although data on calories per egg are generally lacking, egg weight data for S. undulatus support my findings on energy expenditure per offspring. Tinkle and Ballinger (1972) calculated egg weights for four populations of S. undulatus and found that, generally, longer-lived populations produced heavier eggs than shorter-lived populations. In Texas S. undulatus ma-
tured in < 1 yr, produced about 27 eggs/season, and each egg weighed 0.22 g. Conversely, Colorado S. undulatus matured in 2 yr, produced about 16 eggs/season, and each egg weighed 0.45 g. Ballinger
and Clark (1973) showed that relative clutch weight
is a good estimator of relative calories per clutch.
From Ballinger and Clark (1973) and Tinkle and
Ballinger (1972) one can conclude that Colorado S.
undulatus are putting more energy into each egg
than Texas lizards. In fact, if one uses the caloric
and percent water content values estimated by Bal-
linger and Clark for S. undulatus eggs (6,195 cal-
ories/g of egg and 54%), calories per egg can be estimated for each of the S. undulatus populations
studied by Tinkle and Ballinger. Such calculations
yield an estimate of 1,168 calories/egg for Colorado
lizards and 612 calories/egg for Texas lizards. These
figures are surprisingly close to those found for
sagebrush lizards and, prairie lizards respectively, in
the present study.
Data from Tinkle and Hadley (1975) support my
conclusion that the higher reproductive effort by
prairie lizards results in a shorter lifespan. These
authors examined a number of demographic variables
and measures of reproductive effort in 11 lizard spe-
cies and found that the only significant correlation
was an inverse relationship between clutch calories/
body calories and annual adult survivorship. (Inter-
estingly no correlation was noted for clutch weight/
body weight, frequently used as an index of
reproductive effort, and annual adult survivorship).
As seen in Table 4 the clutch calories/body calories
index was 2.53 and 1.45 for prairie lizards and sage-
brush lizards, respectively. Based on Tinkle and
Hadley (1975), prairie lizards would be expected to
have the lower annual adult survivorship that it does.
A comparison of this index with reproductive energy
total energy budget in these authors' study indicates
that clutch calories/body calories may not be a good index of reproductive effort. For example, while
U. stansburiana had a clutch calories/body calories
index of 2.54 the calculated reproductive energy/total energy budget was 19%. In S. graciosus, on the other hand, these values were 1.45 and 24%, respectively. This discrepancy is readily resolved when looking at differences in age-at-maturity. Uta stansburiana is
essentially an annual species and does everything
(matures, reproduces, and dies) in 1 yr, while S. graciosus requires 2 yr to reach maturity. Therefore,
for the comparisons to be meaningful, total energy
budgets of U. stansburiana and S. graciosus would have to be weighted to account for this difference.
This weighting should indicate that clutch calories/ body calories is a reasonable index of reproductive effort.
In summary, evidence in the literature parallels
the findings of this study. There are correlations of
reproductive strategies with lipid levels, percentage of
ingested energy available for metabolism, and energy
expended per offspring. Higher posthibernation lipid
levels are correlated with higher reproductive effort,
while percentage of ingested energy available for
456 W. KENNETH DERICKSON Ecology, Vol. 57, No. 3
metabolism and energy expended per offspring are
inversely related to reproductive effort.
More energy per egg and fitness
Based on demographic data on prairie lizards (Fer-
guson and Bohlen 1976) and sagebrush lizards
(Tinkle 1973) and my data on egg energy values, I
conclude that higher energy content per egg results
in large hatchlings with higher survivorship. Tinkle
and Ballinger (1972) indicated that survivorship of
offspring from larger eggs (greater energy) was
higher. Fourteen percent of Colorado S. undulatus
hatchlings survive to the next spring compared to
5% for Texas hatchlings. In sagebrush lizards first
clutch (4,431 J, or 1,059 cal/egg) hatchlings were
larger (x snout-vent length = 26 mm) than first
clutch (2,833 J, or 677 cal/egg) hatchling prairie
lizards (x snout-vent length - 23 mm). As men-
tioned earlier, this larger size in sagebrush lizards is
adaptive for using a wider array of food items at
lower food levels.
That selection is favoring larger size for lower
food levels is further supported by comparing suc-
cessive clutches of prairie lizards. The mean number
of joules (or calories)/egg was 2,833 (677), 3,071
(734), and 3,573 (854) for the first, second, and
third clutches, respectively. This suggests that the
hatchlings from these clutches should differ in size.
Indeed, hatchlings from the first clutch were signifi-
cantly smaller than those from the third clutch, 22.8
and 24.1 mm (Table 3). If food is critical at all for
prairie lizard hatchlings it should be when the last
clutch is hatching, late August to early September. During this period hatchling density is highest and insect productivity should be declining which should result in less food per individual.
Even within a given clutch larger-sized prairie lizard hatchlings can be shown to have a survival advantage to the next spring. Ferguson and Bohlen (1976) found that third clutch hatchlings with a snout-vent
length > 24 mm had a significantly higher survivor-
ship than those < 24 mm (43% and 21 %, respec- tively). He found no significant difference in sur- vivorship for different-sized hatchlings of the first
clutch. These data suggest that larger hatchling size
is less important in the first clutch than in the third clutch. Besides the advantage to larger hatchlings of using a wider array of food, larger size confers a
dominance advantage during social contests which
could be related to food territories (Rand 1967).
Comparable data are not available on sagebrush lizards. However, since joules (calories) per egg are
significantly higher in the second clutch than in the
first, 5,832 (1,394) and 4,431 (1,059), respectively,
the above data would suggest that second clutch
hatchlings are larger than first clutch hatchlings. If
hatchlings from both clutches are competing for food
in sagebrush lizards, then survivorship should be
higher for the larger hatchlings in both clutches.
Evidence for food competition
A basic assumption underlying the hypotheses
tested in this study was that differences in the degree
of competition for food existed between the two
species. Additionally, it was assumed that the off-
spring of each species were exposed to the same de-
gree of competition for food as the adults. Although
no attempt was made to look at food levels in the
field, data on precipitation levels in the two study
areas and mouthgapes for both species suggest that
these may be valid assumptions.
Analysis of weather data over the past 30 yr for
the two study areas, indicates that precipitation levels
are predictably lower in Utah than in Kansas (Fig.
7). Janzen and Schoener (1968) showed that insect
productivity increases with precipitation levels.
Higher precipitation levels, typically, resulted in
greater insect diversity and abundance. Therefore,
a lower precipitation level suggests that there should
be lower food levels in Utah. Interestingly, there is
a greater density, 208/ha, of sagebrush lizards (Tinkle 1973) than prairie lizards, 42/ha (Ferguson
and Bohlen 1973). Less potential food and a greater
density of lizards strongly suggest that there was less
food per individual for sagebrush lizards than for
prairie lizards with their low density and more po-
tential food. Mid-July until late October should be a
critical time for both the parents and offspring of the two species, since during this time they grow and
store lipids for hibernation. Mean precipitation levels
at this time are 8.71 cm and 2.79 cm for Kansas and
Utah, respectively. Lower precipitation levels should result in less food being available for sagebrush lizard
adults and offspring. In sagebrush lizards, low adult
mortality during the breeding season results in large
numbers of adults surviving (65/ha) during, at
least, the early part of the mid-July to October
period to compete with offspring for food. High
adult mortality in prairie lizards during the breeding
season results in fewer adults (22/ha) to compete
with offspring. Data from Sexton et al. (1972) and
Parker and Pianka (1973) suggested that there is a
great deal of overlap in food resources of adult and
offspring lizards. Competition among the adults and
offspring may further reduce the amount of food
available to sagebrush lizard offspring, although this
may be mitigated by the juveniles utilizing smaller
prey due to their smaller mouths. Differences in pre-
cipitation levels and adult mortality during the
breeding season, therefore, may result in differences
in food availability for the two species after repro-
duction has ceased.
Late Spring 1976 ASPECTS OF REPRODUCTIVE STRATEGIES IN LIZARDS 457
Additional data suggesting food competition was greater for sagebrush lizards were mouthgape and snout-vent length relationships for the two species (Fig. 8). Regression lines showed that given a prairie lizard and a sagebrush lizard of equal size, the latter had a larger mouth. Schoener (1968) found that average prey size and variance of prey size in- creased with head length in Bimini anoles. Presum- ably, longer heads resulted in larger mouths in anoles. Since sagebrush lizards had a larger mouthgape than prairie lizards they should be able to utilize not only larger prey but also a wider range of prey sizes. Species that are food limited should be generalists in their food habits (Emlen 1973). The capacity to handle a wider array of prey sizes suggests that sagebrush lizards may be less specialized in their food habits than prairie lizards.
Finally, Licht (1974) suggested that differences in lipid levels are indicative of different food avail- abilities. Licht was able to increase lipid levels in Puerto Rican anoles by supplemental feeding in the field. Lizards that were not given additional food in the field had much lower lipid levels, suggesting that a limited amount of food is available. My own laboratory experiments (Table 2) demonstrated that with ad libitum feeding, both species deposit lipids faster than field animals. Also, the discrepancy be- tween lipid deposition rates in the field and laboratory were greater in sagebrush lizards (1.3 mg/g LDBW - day) than in prairie lizards (0.9 mg/g LDBW - day). Therefore, these data suggest that a positive correla- tion exists between lipid levels and food availability and that prairie lizards may have more food avail- able than sagebrush lizards.
ACKNOWLEDGMENTS
I thank Donna, Kevin, and Todd for their support and understanding. Their sacrifices made this study possible. The patience, understanding, and considerable talents of my committee members, Drs. Stephen D. Fretwell, G. Richard Marzolf, Christopher C. Smith, and Roger Wein- berg, were instrumental in the development and fulfill- ment of the project. Words cannot express my gratitude to my major professor, Dr. Gary W. Ferguson, for the time and effort he devoted in helping me develop this research problem. Drs. Kenneth Kemp, George Milliken, and Don Yamashita were interested and kind enough to aid with the statistical analyses. Dr. William Klopfen- stein provided the biochemical expertise. Gerard Hod- denbach and his three sons, Billy, Jimmy, and Johnny, provided many of the lizards from Utah. Their interest and love of working with lizards was inspiring. Many graduate and undergraduate students helped collect liz- ards and acted as sounding boards for my ideas. They influenced my thinking and I am indebted to them. I am grateful to Dr. Donald Tinkle, Dr. Tom Poulson, and two anonymous reviewers for their helpful comments and suggestions. This research was supported in part by a NSF Doctoral Dissertation Improvement Grant (No. GB-35156) to me through Dr. Ferguson. The Division of Biology and its faculty supplied much of the materials and equipment needed for this research.
LITERATURE CITED
Avery, R. A. 1971. Estimates of food consumption by the lizard, Lacerta vivipara Jacquin. J. Anim. Ecol. 40:351-365.
Ballinger, R. A., and D. R. Clark, Jr. 1973. Energy content of lizard eggs and the measurement of repro- ductive effort. J. Herpetol. 7:129-132.
Chow, G. C. 1960. Tests of equality between sets of coefficients in two linear regressions. Econometrika 28:591-605.
Church, G. 1962. The reproductive cycles of the Jamaican House Geckos, Cosymbotus platyurus, Hemi- dactylus frenatus, and Peropus mutilatus. Copeia 1962:262-269.
Cole, L. C. 1954. The population consequences of life history phenomenon. Q. Rev. Biol. 29:103-137.
Davis, D. E. 1967. The annual rhythm of fat deposi- tion in woodchucks (Marmota monax). Physiol. Zool. 67:391-402.
Derickson, W. K. 1974. Lipid deposition and utiliza- tion in the Sagebrush Lizard, Sceloporus graciosus: Its significance for reproduction and maintenance. Comp. Biochem. Physiol. 49A:267-272.
Dessauer, H. C. 1955. Seasonal changes in the gross organ composition of the lizard A nolis carolinensis. J. Exp. Zool. 128:1-12.
Emlen, J. M. 1966. The role of time and energy in food preference. Am. Nat. 100:611-617.
. 1973. Ecology: An evolutionary approach. Addison-Wesley Publishing Company, Reading, Massa- chusetts. 493 p.
Ferguson, G. W., and C. H. Bohlen. 1973. The regula- tion of Prairie Swift (Lizard) populations. A progress report. Proc. Third Midwestern Prairie Conf. 3:69- 72.
Ferguson, G. W., and C. H. Bohlen. 1976 (in press). Demographic analysis: A tool for the study of natural selection of behavioral traits. In N. Greenberg and P. McClean [ed.] Conference on brain behavior and evolution in lizards: A colloquium. Adana Press, Washington, D.C.
Folch, J., M. Lees, and G. H. Sloane Stanley. 1957. A simple method for isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226:497- 509.
Gadgil, M., and W. H. Bossert. 1970. Life historical consequences of natural selection. Am. Nat. 104: 1- 25.
Gadgil, M., and 0. T. Solbrig. 1972. The concept of r- and K-selections: Evidence from wild flowers and some theoretical considerations. Am. Nat. 106:14-31.
Gaffney, F. C., and L. C. Fitzpatrick. 1973. Energetics and lipid cycles in the lizard, Cnemidophorus tigris. Copeia 1973:446-452.
Goldberg, S. R. 1973. Ovarian cycles of the Western Fence Lizard, Sceloporus occidentalis. Herpetologica 29:284-288.
Golley, F. B. 1961. Energy values of ecological ma- terials. Ecology 50:517-519.
Gorman, G. C., and P. Licht. 1975. Differences be- tween the reproductive cycles of sympatric Anolis liz- ards on Trinidad. Copeia 1975:332-336.
Hadley, N. F., and W. W. Christie. 1974. The lipid composition and triglyceride structure of eggs and fat bodies of the lizard Sceloporus jarrovi. Comp. Bio- chem. Physiol. 48B:275-284.
Hahn, W. E., and D. W. Tinkle. 1965. Fatbody cycling and experimental evidence for its adaptive significance
458 W. KENNETH DERICKSON Ecology, Vol. 57, No. 3
to ovarian follicle development in the lizard Uta stansburiana. J. Exp. Zool. 158:79-86.
Hairston, N. G., D. W. Tinkle, and H. M. Wilbur. 1970. Natural selection and the parameters of population growth. J. Wildl. Manage. 34:681-690.
Harper, J. L., P. H. Lovell, and K. G. Moore. 1970. The shapes and sizes of seeds. Annu. Rev. Ecol. Syst. 1: 327-356.
Helms, C. W. 1968. Food, fat, and feathers. Am. Zool. 8: 151-167.
Hoddenbach, G. A. 1966. Reproduction in western Texas Cnemidophorus sexlineatus (Sauria: Teiidae).
Copeia 1966:110-113. Horwitz, W. P., P. Chichilo, and Helen Reynolds. 1970.
Methods of analysis. Association of Official Analyti- cal Chemists, Washington, D.C. 1015 p.
Jameson, E. W., Jr., and R. A. Mead. 1964. Seasonal
changes in fatbody fat, water, and basic weight in Citellus lateralis, Eutamias speciosus and E. amoenus. J. of Mammal. 45:359-365.
Janzen, D. H., and T. W. Schoener. 1968. Differences in insect abundance and diversity between wetter and drier sites during a tropical dry season. Ecology 49: 96-110.
Johnson, D. R. 1966. Diet and estimated energy as- similation of three Colorado lizards. Am. Midl. Nat. 76:504-509.
King, J. R. 1970. Photoregulation of food intake and fat metabolism in relation to avian sexual cycles, p. 365-385. In J. Benoit and I. Assenmacher [ed.] La Photoregulation de la reproduction chez les oiseaux et les mammiferes. Editions du C. N. R. S. Paris.
Licht, P. 1974. Response of Anolis lizards to food supplementation in nature. Copeia 1974:215-221.
Licht, P., and G. C. Gorman. 1970. Reproductive and
fat cycles in Caribbean Anolis lizards. Univ. Calif. Publ. Zool. 95:1-52.
MacArthur, R. H., and E. R. Pianka. 1966. On opti- mal use of a patch environment. Am. Nat. 100:603- 609.
MacArthur, R. H., and E. 0. Wilson. 1967. The theory of island biogeography. Monographs in Popu- lation Biology, Princeton, New Jersey. 203 p.
Mueller, C. F. 1969. Temperature and energy char- acteristics of the Sagebrush Lizard (Sceloporus gracio- sus) in Yellowstone National Park. Copeia 1969: 153-160.
1970. Energy utilization in the lizards Sceloporus graciosus and S. occidentalis. J. Herpetol. 4:131-134.
Nikolskii, G. V. 1969. Theory of fish population dynamics as the biological background for rational exploration and management of fishery resources. Oliver and Boyd, Edinburgh. 323 p.
Odum, E. P. 1960. Lipid deposition in nocturnal mi- grant birds. Proc. 12th Inst. Omithol. Congr. p. 563- 576.
Paine, R. T. 1971. The measurement and application
of the calorie to ecological problems. Annu. Rev. Ecol. Syst. 2:145-164.
Parker, W. S., and E. R. Pianka. 1973. Notes on the ecology of the iguanid lizard Sceloporus magister. Herpetologica 29:143-151.
Pianka, E. R. 1970. On r and K selection. Am. Nat. 104:592-597.
1972. r and K selection or b and d selection? Am. Nat. 106:581-588.
Pough, F. H. 1973. Lizard energetics and diet. Ecol- ogy 54:837-844.
Rand, A. S. 1967. Ecological and social organization in the iguanid lizard Anolis lineatopus. Proc. U.S. Natl. Mus. 122:1-79.
Ruibal, R., R. Philibosian, and J. L. Adkins. 1972. Re- productive cycle and growth in the lizard Anolis acutus. Copeia 1972:509-518.
Sawicka-Kapusta, K. 1968. Annual fat cycle of field mice, Apodemnes flavicollis (Melchior, 1834). Acta Theriol. 19:329-339.
Schoener, T. W. 1968. The Anolis lizards of Bimini: Resource partitioning in a complex fauna. Ecology 49:704-726.
1971. Theory of feeding strategies. Annu. Rev. Ecol. Syst. 2:369-404.
Sexton, 0. J., Joan Bauman, and E. Ortleb. 1972. Sea- sonal food habits of Anolis limifrons. Ecology 53: 182-186.
Smith, C. C., and S. D. Fretwell. 1974. The optimal balance between size and number of offspring. Am. Nat. 108:499-506.
Smith, H. M. 1946. Handbook of lizards. Comstock Publishing Company, Ithaca, New York. 557 p.
Smith, R. E. 1968. Studies on reproduction in Costa Rican Ameiva festiva and Ameiva quadrilineata (Sauria: Teiidae). Copeia 1968:236-239.
Stebbins, R. C., and H. B. Robinson. 1946. Further analysis of a population of the lizard Sceloporus graciosus gracilis. Univ. Calif. Publ. Zool. 48:149- 168.
Telford, S. R., Jr. 1970. Seasonal fluctuations in liver and fatbody weights of the Japanese lacertid Takydro- mus tachydromoides. Copeia 1970:681-689.
Tinkle, D. W. 1969. The concept of reproductive ef- fort and its relation to the evolution of life histories of lizards. Am. Nat. 103:501-516.
. 1973. A population analysis of the Sagebrush Lizard, Sceloporus graciosus in southern Utah. Copeia 1973:284-296.
Tinkle, D. W., and R. A. Ballinger. 1972. Sceloporus undulatus: A study of the intraspecific comparative demography of a lizard. Ecology 53:570-584.
Tinkle, D. W., and N. F. Hadley. 1975. Lizard repro- ductive effort: Caloric estimates and comments on its evolution. Ecology 56:427-434.
Tinkle, D. W., H. M. Wilbur, and S. G. Tilley. 1970. Evolutionary strategies in lizard reproduction. Evolu- tion 24:55-74.