METABOLIC RATES IN SPIDERS
JOHN FRANCIS ANDERSON
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF TIHE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
I am deeply indebted to Dr. James L. Nation, under whose direction
the present research was conducted.
Dr. Robert M. DeWitt must be thanked for his liberal support in
providing the necessary supplies and equipment.
Grateful acknowledgement is given to Dr. Frank G. Nordlie, Dr.
Brian K. McNab, Dr. John D. McCrone, Dr. James L. Nation, and Mr.
Thomas Krakauer who gave most generously of their time in evaluating
some of my ideas.
My thanks go to Dr. Martin H. Muma, Dr. John D. McCrone, and Mr.
Cole Benton for supplying certain animals which otherwise would not
have been available. For aid in solving various statistical problems,
I am indebted to Miss Mary E. Glenn. Dr. Rodger Mitchell must be
thanked for his assistance in conducting the histological studies.
Financial support through appointment to research and teaching
positions was provided by Dr. Frank G. Nordlie and Dr. Lewis Berner
to whom I am grateful.
Dr. Nation, Dr. Nordlie, and Dr. McCrone must again be thanked
along with other members of my graduate supervisory committee, Dr.
Thomas J. Walker and Dr. Howard K. Wallace, who not only critically
reviewed the dissertation, but also provided support in solving the
many small but important problems that occurred during my tenure as a
Finally thanks are also due to Mrs. Lillian Ingenlath for typing
TABLE OF CONTENTS
ACKNOWLEDGEMENTS. . . . . . . . . ... ... ii
LIST OF TABLES. ............ .. . ... ..... . iv
LIST OF FIGURES . . . . . . . . ... . . v
INTRODUCTION . . . . . . . . . . . I
MATERIALS AND METHODS ............... . .. 5
RESULTS . . . . . . . . ... .... . . .. 14
Weight Changes . . . . . . . . ... . . 14
Temporal Variation in Oxygen Consumption . . ... 14
Loss of Temporal Variation in Oxygen Consumption . 19
Standard Metabolic Rates ................ .20
Respiratory Quotient .................. 23
Breathing Systems. ................ . . 23
Exoskeleton Weight, Metabolically Active Tissue
Weight, and Total Weight.... . . . . . 23
Oxygen Consumption and Temperature . . . . .... 23
DISCUSSION . . . ........ .... . ..... . 37
LITERATURE CITED. ............... ..... 64
BIOGRAPHICAL SKETCH ............... . . 67
LIST OF TABLES
I LIST OF EXPERIMENTAL ANIMALS. . . . . . . . 6
2 WEIGHT CHANGES IN SPIDERS . . . . . . .... 14
3 AVERAGE OXYGEN CONSUMPTION OF ARACHNIDS AT 20 C . . 20
4 OXYGEN CONSUMPTION OF INDIVIDUAL ARACHNIDS AT
20 C. . . . . . . . . ... . . . . 21
5 CALCULATED SLOPES AND INTERCEPTS OF THE EQUATION
LOG METABOLISM = LOG K + n LOG WT . . . .... .22
6 RESPIRATORY GAS EXCHANGE SYSTEMS OF SPIDERS ...... 24
7 TOTAL WEIGHTS, EXOSKELETON WEIGHTS, AND ACTIVE
TISSUE WEIGHTS. . . . . . . . . ... ... 26
8 OXYGEN CONSUMPTION AT DIFFERENT TEMPERATURES. . . ... 29
9 AVERAGE WEIGHT, METABOLIC RATE AND BOOK-LUNG
SURFACE AREA IN ADULT SPIDERS . . . . . .... .50
10 CALCULATED RELATIONSHIPS BETWEEN EXOSKELETON
WEIGHTS AND METABOLICALLY ACTIVE TISSUE WEIGHTS
TO TOTAL WEIGHTS. . . . . . . .... ... . 57
LIST OF FIGURES
1 Oxygen consumption over 24 hours. Each point
represents the mean of N determinations. The
time interval indicated by the black stripe
represents the 12 hour dark period. (a) L.
lenta, N = 6; (b) f. hibernalis, N = 14. . . . .. 16
2 Oxygen consumption over 24 hours. Each point
represents the mean of N determinations. The
time interval indicated by the black stripe
represents the 12 hour dark period. (a) P.
regius, N = 6; (b) T. rufipes, N = 5; (c) A.
tepidariorum, N = 5. . . . . . . ... .. 18
3 Oxygen consumption of P. reaius versus
temperature. e 10 C acclimation; o 20 C
acclimation; e 30 C acclimation. Acclimation
time exposures are indicated in Table 8. . . . ... 31
4 Oxygen consumption at 30 C. Vertical lines
represent 95% confidence intervals of the
sample means. Spiders were exposed to 20 C
for three weeks prior to the start of this
experiment. (a) P. reqius; (b) F. hibernalis. . .. 34
5 Oxygen consumption at 10 C. Vertical lines
represent 95% confidence intervals of the
sample means. Spiders were exposed to 20 C
for three weeks prior to the start of this
experiment. (a) F. hibernalis; (b) L. lenta . . .. 36
6 Relationship between metabolic rate and
weight. (a) pooled regression line for all
arachnids; (b) theraphosid species (adults
and immatures); (c) F. hibernalis (adults);
(d) L. lenta (adults); (e) L. lenta (immatures);
(f) P. regius (adults); (g) P. reius (immatures);
(h) T. rufipes (adults); (i) P. otiosus (adults);
(j) T. sisyphoides (adult); (k) E. bilobatus (adults)
(1) A. tepidariorum (adults); (m) C. vittatus (adults)
(n) U. audouini (immature); (o) T. marginemaculata
(adults); and (p) C. hentzi (adults) . . . ... .41
7 Relationship between metabolic rate and weight.
(a) Hemmingsen's (1960) poikilotherm line;
(b) pooled arachnid line. Vertical lines
represent 95% confidence intervals . . . . .. 47
8 Relationship between book-lung surface area
and oxygen consumption in spiders. (a) L.
lenta; (b) P. reqius; (c) F. hibernalis;
(d) T. sisyphoides; (e) A. tepidariorum;
and (f) T. rufipes . . . . . . . . ... .52
9 Deviations of actual versus calculated values
of oxygen consumption and respiratory surface
area. Abscissa, surface area deviations.
Ordinate, oxygen consumption deviations.
(a) A. tepidariorum; (b) L. lenta; (c) P.
reaius; (d) T. sisyphoides; (e) T. rufipes;
and (f) F. hibernalis. . . . . . . . . 55
10 Relationship between temperature and standard
metabolism in P. reqius and F. hibernalis.
(a) minimal required energy expenditure at
20 C; (b) minimal required energy expenditure
at 30 C; (c) compensation at 30 C in P. reqius .... 62
The diversity of spiders is well known in terms of anatomy, be-
havior, and ecology. These aspects have been documented by Petrunke-
vitch (1933), Gertsch (1949), and Millot (1949). Although Anderson
(1966) studied the excreta of many different spiders, most other phys-
iological studies have dealt with a relatively small number of species.
This situation has resulted in a lack of information pertaining to
physiological diversity and,as such, precluded a fuller understanding
of the biology of these arachnids. I thought that respiration, as
measured by oxygen consumption, would be the most productive variable
to analyze in this regard, not only because of its ease of measurement,
but because it reflects the total energy requirement for all the meta-
bolic processes. This latter point was made by Benedict (1938), and
is supported by the wealth of literature where specific physiological
processes were studied using some aspect of respiration.
The major aim of this study was to determine whether interspecific
differences in oxygen consumption exist and, if so, to uncover some of
the causative factors in order to explore the consequences of these
differences in relation to the biology of spiders.
Interspecific differences have been detected in a number of
arachnids: they have been discovered in oribatid mites by Berthet
(1963); in harvestmen by Phillipson (1962); and in scorpions by Dresco-
Derouet (1964). In regard to spiders, Derouet (1953) found that a
cavernicolous species had a lower metabolic rate than one which nor-
mally lives outside caves.
A second aim of this study was to determine whether being a spider
is reflected in an energetic sense by comparing the metabolic rates of
spiders with those of other poikilotherms. There is some difficulty
in making these comparisons; the weights of the groups being compared
often differ, and the relationship between weight and metabolism is
not one of direct proportionality. While larger animals have a greater
total rate of metabolism than smaller ones, the converse is true when
metabolism is put on a per unit weight basis. The relationship be-
tween weight and metabolism has generally been represented by the
metabolism = k weight
log metabolism = log k + n log weight
The logarithmic form of the equation is useful since most of the values
of n do not equal one, hence an arithmetic plot results in a curved
line. A double logarithmic plot of the equation for any value of n
does give a straight line. Consequently, to compare metabolic rates,
values of n and k are determined from experimental data. The equations
represented by these constants can then be compared with one another
as well as with standard curves. The latter have been constructed by
pooling data obtained from a wide variety of organisms. Examples of
such standards are Hemmingsen's (1960) unicellular, poikilotherm, and
homeotherm metabolism-weight regression lines.
I also thought it valuable to consider factors which might be
responsible for the size of n and k. Hemmingsen (1960) suggested that
the differences between levels of metabolism of poikilotherms and
homeotherms, i.e., different values of k, may be the result of an
increase in respiratory surface area in the latter group. Tenney and
Remmers (1963) demonstrated a relationship between lung surface area
and metabolic rate in a group of mammals ranging from bats to whales.
In poikilotherms, both Brown (1957) and Hemmingsen (1960) cite data
indicating that active fish have larger gill surface areas than sluggish
fish of the same weight. Whitford and Hutchison (1967) demonstrated
that above 15 C, lunged salamanders have a higher metabolic rate than
lungless species of the same weights. These findings indicated the
desirability of making interspecific comparisons of the relationship
between surface areas of book-lungs and metabolic rates in spiders.
In addition, I thought it of value to investigate metabolic rates
of those arachnids having only book-lungs or only trachea. The
Solifugae, as most insects, utilize only trachea while the Amblypygi
utilize only book-lungs as breathing organs. Thus, knowledge of
metabolic rates of solifugids and amblypygids may make comparison of
metabolic rates between spiders and insects more meaningful.
In the latest comprehensive review of the subject, Hemmingsen
(1960) claims that n, the slope of the metabolism-weight regression
equation, equals 0.75 and is quite uniform when large weight and taxo-
nomic ranges are considered. He claims the value of 0.75 has evolved
through orthoselection as the result of the struggle between propor-
tionality of metabolism to body weight and to surface functions.
Prosser and Brown (1961) and Hoar (1966) presented a more specific
hypothesis to explain the size of n. They suggest metabolism is not
directly proportional to weight because as animals get larger, a dis-
proportionate increase in supportive tissue is required. This hypoth-
esis assumes that supportive tissues such as the skeleton have low
rates of metabolism. I decided to zest this hypothesis by assuming
that the exoskeleton constitutes the bulk of tissue having a low meta-
bolic rate in spiders.
As a final consideration, the effect of temperature on metabolic
rate was studied in order to better evaluate any interspecific differ-
ences in metabolismo. !n general the rates of metabolism of poikilotherms
are governed by a logarithmic law relating the velocities of the various
component reactions to temperature. However, strict adherance to this
relationship is not always followed by poikilotherms. Some variations
in the metabolic response to temperature include the temperature com-
pensation response in individuals when exposed to new temperatures as
discussed by Bullock (1955), and the relatively new findings of Newel!
(1966) and Newell and Northcroft (1967, indicating temperature inde-
pendence to rapid temperature changes in quiescent animals.
With these possibilities in mind, a series of experiments was
designed to answer questions about the ability of spiders to remain
independent of rapid temperature fluctuations; the ability, degree,
and rate of temperature compensation; and the effect of previous thermal
history with regard to metabolic rate.
MATERIALS AND METHODS
To increase the probability of detecting interspecific physiolog-
ical differences, a diverse group of spiders was selected for study.
Two species of scorpions were included to aid in comparing spider
metabolism with that reported for scorpions by Dresco-Derouet (1964).
Adult females were used for the majority of observations as the relative
scarcity, short life span, and relatively high activity patterns of
males made them unsuitable for study. Immatures of a few species were
investigated. The species selected and site of their collection are
listed in Table I.
In general, metabolic rates of poikilotherms are measured under
conditions designed to obtain the lowest rate compatible with life
when all organs are at minimal levels of activity. Measurements so
obtained are termed standard metabolic rates. The conditions used to
maintain the arachnids in the laboratory and the conditions under which
measurements of standard metabolism were made are described in the
The arachnids were housed individually in glass containers of
appropriate size. Sheets of balsa wood were added to the cages of the
web building species and the salticids to provide a framework for webs
or the silken retreats of the latter. The cages of the other arachnids
were partially filled with moist sand.
lhe arachnids were kept at a temperature of 20 C except during the
LIST OF EXPERIMENTAL ANIMALS
Family Species Collection Site
Ummidia audouini (Lucas)
Lycosa lenta Hentz
Phidipppus reqius (C.L. Koch)
Phidippus otiosus Hentz
Filistata hibernalis (Hentz)b
Theridion rufipes Lucas
Centruroides hentzi (Banks)
12 miles north of
Mexico on Highway
5 miles west of
Mexico on Highway
TABLE 1 (Continued)
Centruroides vittatus (Say)
Eremochelis bilobatus (Muma)
Monroe Co., Florida
aThe confused state of taxonomy of this group prevented species deter-
bLehtinen (1967) places this species in the genus Kukulcania.
experiments dealing with the effects of temperature change. This
temperature was selected because it approximates the average temperature
where most of the arachnids were collected. It is also a temperature
commonly used in measuring metabolic rates of poikilotherms; thus it
facilitates the comparison of data with those of other workers.
Relative humidity was kept near 100% by the addition of a few drops of
water to the cages at frequent intervals. In addition, a shallow tray
of water was kept in the incubator. A high relative humidity was
maintained since it was discovered that certain lycosid spiders tended
to reduce their oxygen consumption when the humidity was lowered to
10%. In essence, these conditions were used to prevent undue stress on
the animals. A 12 hr light and dark cycle was maintained in the in-
cubator with illumination provided by a 40 w bulb or a 40 w fluorescent
tube. The arachnids were fed various types of living insects once a
week. The purpose of the feeding schedule was to maintain constant
body weights. The efficacy of the feeding schedule was tested by
monitoring individual weights. Oxygen consumption for the purpose of
obtaining standard metabolic rates was measured at 20 C, 100% relative
humidity, and at a point of minimal activity. ?hillipson (1962) dem-
onstrated temporal variations in oxygen consumption in harvestmen.
Phanuel (1967) and Dresco-Derouet (1961) detected patterns in oxygen
consumption corresponding to activity cycles in certain salticid
spiders and scorpions, respectively. These results pointed to a
potential problem of defining the time of day when metabolism should
be measured. Consequently the oxygen consumption of a number of species
was measured over a 24 hr period at 20 C, 1007, relative humidity, and
a 12 hr light and dark cycle to find the time of day when the metabolic
rates were consistent with the definition of standard metabolism.
Once determined for each species, this time period was used for all
Since my definition of standard metabolism depended on the exis-
tence and maintenance of a daily rhythm of activity where I could measure
oxygen consumption at a time whe'n.the arachnids were not active, I
thought it valuable to set up a control experiment to detect possible
sources of error if the acclimation conditions inadvertently produced
a loss of rhythm. Individuals of L. lenta, a species which from field
observations and preliminary studies of oxygen consumption was known
to exhibit temporal variations in activity and oxygen consumption,
were kept under constant darkness at 20 C and 100% relative humidity
for three weeks. The oxygen consumption of this experimental group
of L. lenta was then measured in the evening, a time when they are
normally active in nature. The results were compared with standard
metabolic rates of a control group of L. lenta that had been kept at
20 C and 100% relative humidity but under a 12 hr light and dark cycle
to see if loss of the temporal pattern of oxygen consumption occurred,
and also to see whether the rates of oxygen consumption obtained for
the experimental group were comparable to standard metabolic rates for
Observations were made on individuals after they were exposed to
20 C, 100'/ relative humidity, and the 12 hr light and dark cycle for
three weeks. Metabolic rates were not measured on any individual
until six days after their last feeding to preclude variations in
metabolic rates due to the absorption of food from the gut.
Actual oxygen consumption was measured using a Gilson differential
respirometer. Warburg flasks of about 15 ml capacity were used to
house the arachnids in most instances. Some of the larger lycosids
and all the theraphosids required the use of 125 and 200 ml flasks.
About 0.2 ml of a 10% (w/v) KOH solution was placed in the center wells
of the flasks to absorb the expired carbon dioxide. Granular soda
lime, (a mixture of NaOH and Ca(OH)2), was used for this purpose in the
case of the larger flasks. About an inch of the soda lime was layered
at the bottom of the flasks and covered with several thicknesses of
plastic screening to prevent contact with the animals. One-half ml of
water was placed in the side arms to produce a water saturated atmos-
phere within the flasks. For those spiders whose oxygen consumption
was measured under conditions simulating daylight, 75% of the flask
was covered with black plastic tape to minimize the effect of bright
light. All measurements were converted to ul of dry gas at 760 mm Hg
and 0 C.
An estimate of the respiratory quotient of the arachnids was re-
quired to compare oxygen consumption data with data of other workers
whose results are reported in kilocalories. This estimate was obtained
using P. reqius. Twenty-five specimens were.placed individually in
Warburg flasks as usual except that no KOH was added to the center
well. A 30 minute reading was made. Since no KOH was present, any
changes in the level of the manometric fluid reflected both oxygen
consumption and carbon dioxide production. A second 30 minute reading
was made with KOH in the center well. As the expired carbon dioxide
was absorbed by the KOH, this second measurement reflected only oxygen
consumption. By assuming oxygen consumption was constant for the two
30 minute intervals, I calculated carbon dioxide production and sub-
sequently the resoi~-,;orv co"ier'.
Estimates of book-lung surface area were made from measurements
of scaled drawings of serial cross-sections of spider abdomens. The
histological procedures used were described by Mitchell (1964). The
sections were 8 u thick and stained with a Chlorazol black E saturated
solution of 70% (v/v) ethyl alcohol. Drawings of sections of a book-
lung were made with the aid of a split image drawing tube. Every
fifth or tenth section was drawn depending on the size of the spider,
hence estimates of surface area and volume were made from 10 or 20%
of the total number of sections. The volume of a book-lung was esti-
mated from measurements made using a polar planimeter, while total
respiratory surface area was estimated from measurements of individual
lamellae in the drawn sections. The lamellae refer to the leaves of
the book-lung as they appear in cross-section. Since the spiders
investigated have a pair of book-lungs, both volume and surface area
values were doubled to provide estimates for the whole spider.
The extent of development of the tracheal systems was obtained
by examination of specimens after they had been boiled in a 10% (w/v)
solution of KOH.
The relationship between exoskeleton weight and total weight was
calculated from data obtained as follows. Live spiders were weighed
one week after they had fed. The spiders were killed and the viscera
were removed from the exoskeleton by scraping with a scalpel. The
contents of the legs were removed by rolling a glass cylinder down the
length of the legs so as to force extrusion of their contents from the
detached proximal ends. The exoskeleton was dried at 105 C until a
constant weight was obtained.
To study effects of temperature, members of a species were kept
at 20 C, 100% relative humidity, and a 12 hr light and dark cycle for
three weeks. The metabolic rate was measured at 20 C. The temperature
was then changed to 30 C and oxygen consumption monitored until a
stable set of readings was obtained. This reading represents the
metabolic rate at 30 C after acclimation to 20 C. The spiders were
kept at 30 C for six days with oxygen consumption measured daily to
detect temperature compensation effects. On the sixth day, after
completion of the 30 C reading, the temperature was changed first to
20 C and then to 10 C. Oxygen consumption was measured at both tem-
peratures. The rates obtained at 30 C, 20 C, and 10 C on the sixth
day were assumed to represent metabolic rates at the respective tem-
peratures after acclimation to 30 C. The actual time required to
effect a temperature change of 10 C varied between 15 to 20 minutes.
Once the new temperature had been reached, one to three hours was
required to obtain a stable metabolic rate at the new temperature.
The entire procedure was repeated to study acclimation at a
temperature of 10 C. Individuals of a species were acclimated to 20 C,
100% relative humidity, and a 12 hr light and dark cycle for three
weeks. At the conclusion of the acclimation period, metabolic rates
of these individuals were measured at 20 C. The temperature was
changed to 10 C and oxygen consumption monitored until a stable meta-
bolic rate was obtained. This reading represents the metabolic rate
at 10 C after acclimation to 20 C. The spiders were kept at 10 C for
eight days with oxygen consumption measured daily. On the eighth day,
after completion of the 10 C reading, the temperature was changed to
20 C and then to 30 C. The spiders were kept at these two temperatures
until a stable metabolic rate was obtained. A time interval of one to
three hours was required at each temperature. The rates obtained on
the last day at 10 C, 20 C, and 30 C were assumed to represent metabolic
rates in response to acclimation to 10 C.
The temperature range of 10 to 30 C was selected as one the animals
might normally encounter in nature.
Weight Changes: The results obtained from weight measurements are
summarized in Table 2.
WEIGHT CHANGES IN SPIDERS
Species Number of Avg Original Avg Final Interval Between
Individuals Wt t SE Wt + SE Measurements
Measured (mg) (mg) Minimum Maximum
F. hibernalis 23 594 27 537 t 23 1 week 6 months
L. lenta 13 925 + 54 943 + 75 2 weeks 6 months
P. reqius 24 577 + 23 555 + 24 1 week 2 months
T. rufipes 12 28 2 28 + 2 2 weeks 1 month
aOnly mature females were used in this part of the study.
bNo initial weighing were made until the spiders had been held under
acclimation conditions for at least three weeks.
There were no statistically significant differences between average
initial and final weights in the species tested. On the basis of this
evidence I assumed that the feeding procedures were adequate to maintain
a constant body weight in all the experimental animals.
Temporal Variation in Oxygen Consumption: The average oxygen con-
sumption over a 24 hr period is shown in Figures 1 and 2 for individual
species. Sample sizes are indicated in each figure. Temporal variation
Figure 1. Oxygen consumption over 24 hours. Each point represents
the mean of N determinations. The time interval indicated
by the black stripe represents the 12 hour dark period.
(a) L. lenta, N = 6; (b) F. hibernalis, N = 14.
0 __ __ __ _
TIME OF DAY
* 0 0 0 0
* 0 *
0700 1 00 0700
TIME OF DAY
Oxygen consumption over 24 hours. Each point represents the
mean of N determinations. The time interval indicated by the
black stripe represents the 12 hour dark period.
(a) P. reqius, N = 6; (b) T. rufipes, N = 5; (c) A. tepidar-
iorum, N = 5.
S0 0 0
0O 0 0
TIME OF DAY
0 0 0 0 *
TIME OF DAY
1900 07 00
TIME OF DAY
=-- 13 M
was evident for L. lenta and P. reqius with the highest rates of oxygen
consumption occurring at a time when each species is active in nature.
Subsequent measurements for obtaining standard metabolic rates were
made in the evening for P. reqius and during the daytime for L. lenta
when their respective rates reached minimal levels. Temporal variation
was not detected in the other species. Their metabolic rates appear
to be slightly higher at night, but the variation between the hourly
readings and the relatively small magnitude of the rise at night pre-
vented making a firm decision about the existence of a daily pattern
of oxygen consumption in these species. Measurements of oxygen con-
sumption over a 24 hr period were not made on the other arachnids.
I felt the close correspondence between the patterns of oxygen con-
sumption with the normal activity rhythms of L. lenta and P. reqius,
as well as a knowledge of the activity cycles of the other arachnids
studied, justified measurement of oxygen consumption of the latter
group during that part of their activity cycle when physical activity
was known to be minimal.
Loss of Temporal Variation in Oxygen Consumption: The control
group of 17 individuals of L. lenta that had been exposed to an alter-
nating 12 hr light and dark cycle had a mean standard metabolic rate
of 81 ul 02/gm/hr with a SE of t 3. The experimental group of 11
individuals, whose metabolic rates were measured during the evening
after they had been exposed to constant darkness for three weeks, had
a mean rate of 81 ul 02/gm/hr with a SE of + 6. The correspondence of
the metabolic rates of the control group measured during the daylight
and the experimental group measured during the evening when members of
this species are normally active suggests that constant darkness extin-
guished the temporal variation in oxygen consumption. It also suggests
that standard metabolic rates would be obtained even though acclimation
conditions extinguished any metabolic rhythm.
Standard Metabolic Rates: The standard metabolic rates of the
various arachnids tested are listed in Tables 3 and 4. Individual data
are reported in Table 4 since small sample sizes did not make it feasi-
ble to pool data in certain species. The relationship between weight
and metabolic rate for individual species and various groupings of
species has been computed, and the parameters associated with the re-
gression equation that describes the relationship,as well as correlation
coefficients,are reported in Table 5.
AVERAGE OXYGEN CONSUMPTION OF ARACHNIDS AT 20 C
Species Number of Avg Wt + SE Avg Metabolic Avg Wt Specific
Individuals (mg) Rate + SE Metabolism + SE
Measured (ul 02/hr) (ul 02/gm/hr)
F. hibernalis 81 571 15 29 + 2 50 t 2
P. reqius 78 568 35 54 + 2 93 1 4
P. otiosus 13 337 + 30 48 + 5 147 + 19
L. lenta 80 970 27 90 + 5 94 + 5
A. tepidariorum 7 73 11 25 3 356 33
T. rufipes 51 25 1 5 + 1 201 + 11
C. hentzi 15 212 + 18 9 I 43 + 6
aOnly mature females were used in this part of the study.
OXYGEN CONSUMPTION OF INDIVIDUAL ARACHNIDS AT 20 C
Species Specimen Wt Metabolic Wt Specific
(mg) Rate Metabolism
(ul 02/hr) (ul 02/gm/hr)
M series adults
M series immatures
H series adult
H series immatures
G series adult
G series immatures
co r- .D
r- \. co
C- C C
- C0 -T f CO -
- Co) m r- -
- c\ o
I I I
Co C 0
ja *- -
*- 3 3
Respiratory Quotient: The average respiratory quotient of 25
individuals of P. regius was 1.02 with SE equal to + 0.04. On the
basis of these measurements, the conversion ratio of 5.05 kilocalories
per liter of oxygen was used to convert oxygen consumption values to
kilocalories for all the species studied. The conversion made possible
comparison of the metabolic rates of the arachnids studied with the
data reported by Hemmingsen (1960).
Breathing Systems: Surface area and volume estimates of book-
lungs for individuals, as well as a description of the degree of devel-
opment of the tracheal system on the species level are given in Table 6.
Exoskeleton Weight, Metabolically Active Tissue Weight, and Total
Weight: Individual exoskeleton weights, total weights, and estimates
of weights of metabolically active tissue are reported in Table 7. The
estimates of active tissue weights were obtained by subtracting exo-
skeleton weights from total weights.
Oxygen Consumption and Temperature: The rates of oxygen consump-
tion obtained at different temperatures are listed in Table 8. Figure
3 provides a graphic illustration of the relationship between oxygen
consumption and temperature in P_. reqius.
The sensitivity of metabolism to temperature change is evident
when rates of oxygen consumption at different temperatures within the
same temperature acclimation regimen are compared.
The existence and degree of temperature compensation of individ-
uals of a species at a relatively high temperature of 30 C can be
determined by comparison of the values obtained at 30 C to acclimation
temperatures of 20 and 30 C. All the species tested, except F. hibernalis,
show an appreciable reduction in oxygen consumption during a six day
ULA\ C J CO
Lfr\ r Ltr\
- o- O -
Lr\ r C" -3r \10
C; ** C;c ;CI=
In I~x ~
i (NI cT\ LA
- C -
co C E
c l- h:
c a) a o
a E E --
E 0 0 04
O- .-- a U
0 Ln -- c
> a) CO > D a)
0U v CL a)
> 0 0 O c
a > Fo c-
0 U -0 0 >. 0
C> Q.- .(0 in
cr n LO
"~oc 4 0 IA '
.-E r r
0 C -
LIi E 000 0-- --
a) a) m
V) <1 <
TOTAL WEIGHTS, EXOSKELETON WEIGHTS, AND ACTIVE TISSUE WEIGHTS
Species Total Wt Exoskeleton Wt Wt of Active Tissue
(mg) (mg) (Total Wt minus Wt
TABLE 7 (Continued)
Species Total Wt Exoskeleton Wt Wt of Active Tissue
(mg) (mg) (Total Wt minus Wt
TABLE 7 (Continued)
Species Total Wt Exoskeleton Wt Wt of Active Tissue
(mg) (mg) (Total Wt minus Wt
T. rufipes 40 2 38
37 2 35
15 1 14
32 2 30
18 1 17
29 2 27
30 1 29
23 2 21
23 2 21
24 1 23
17 2 15
f2 Lf\ -3:00 Wr- WO L\ P%0
Coo Lr Loo tLt o oD m m0 -0oo
\.O 0o 0o r r- L% Lt> L t Ci O
E I I I I l I I I I I 0
S00 on ncd CN4 C4- CMJ4f
S N -n C- - - n
S--- a0 .D'D I I
S I I I I I I I I I I -
< > 00N \cO4D L(N00 \\ Qe\ L 0
o 06 >-
f) I UI C- -C
Lu ( O 4- N 44 -en- C 0)
- cco n r 0-- ---) CL E
I CD 0
L > 11 0)0
-1 0 I I C N
< < o-'e- -N CNJ l U) L
Q- C4 --M - 0 )
0 0O-c rIOOO -1OU C -O
LLJ u ca - ^- ^-^ -^) 0
0- NNN n N- v 0
> I I I I I I I I
4-U >-0> >.0 > 0 > 0 0
2 0 0)
S0 U -a 0-
= u 0 > Do
S-0- 000 00 00 00 0 00 m C \-o '
-0 -N I I n
_c -a E
-- m U -
2L > U I I 1 1 1 1|- O,
-a 0 C 0M C> C (C M C M (D | | -o 0 U
*- m E 0 0 v
E0 D 00
V) 4a 0 4-
I- 1n CI CLV-
/ Figure 3.
Oxygen consumption of P. reqius versus temperature.
* 10 C acclimation; o 20 C acclimation; 0 30 C
acclimation. Acclimation time exposures are in-
dicated in Table 8.
exposure to 30 C. Statistically significant differences at the 5%
level are apparent in the case of P. regius and T. rufipes. The effects
of 30 C over time in the metabolic rates of P. reqius and F. hibernalis
are indicated in Figure 4. Comparison of the oxygen consumption rates
obtained at 10 C after acclimation to 10 and 20 C does not reveal any
statistically significant increase in oxygen consumption during an
eight day exposure to 10 C. In fact F. hibernalis shows a decrease in
oxygen consumption indicating a type of inverse compensation. The
time course of measurements at 10 C is indicated in Figure 5 for F.
hibernalis and L. lenta.
Oxygen consumption at 30 C. Vertical lines represent 95%
confidence intervals of the sample means. Spiders were
exposed to 20 C for three weeks prior to the start of this
experiment. (a) P. reqius; (b) F. hibernalis.
B I I
Figure 5. Oxygen consumption at 10 C. Vertical lines represent 95%
confidence intervals of the sample means. Spiders were
exposed to 20 C for three weeks prior to the start of this
experiment. (a) F. hibernalis; (b) L. lenta.
o0 1 24 5 6
Although the study of variation in oxygen consumption over a 24
hr period was conducted for the practical purpose of determining when
this measure represented standard metabolic rate, the results have some
Of the five species tested, the two hunting spiders, L. lenta and
P. reggius, displayed a definite temporal variation in oxygen consump-
tion. Since these observations were recorded over a single 24 hr
period, some objection might be made to calling this variation a
circadian rhythm. However, studies of oxygen consumption in harvest-
men by Phillipson (1962) and those on locomotory activity patterns ;n
salticid spiders reported by Phanuel (1967), where measurements were
recorded over two or more 24 hr cycles, support the conclusion that
the two spiders have a definite rhythm of oxygen consumption which
corresponds to periods of activity in nature. It is doubtful, however,
that the measured oxygen consumption of these hunting spiders during
their active locomotory phases accurately reflects the oxygen consump-
tion maintained during such phases in nature. While the high rates
obtained in the laboratory probably correspond to the time these spiders
are active in the field, the magnitude of such measurements may be a
reflection of the animals struggling within the confines of a small
Cloudsley-Thompson (1960) suggests the adaptive functions of these
rhythms may partially reside in the avoidance of harsh physical factors.
This function is probably of some significance in the case of nocturnal
species such as L. lenta. However, the biotic function of synchronizing
activity of spiders with the time of greatest prey availability cannot
be excluded as a possibility.
A definite temporal pattern in oxygen consumption is not apparent
from the results graphed in Figures Ib, 2b, and 2c. The species in-
volved are all web builders; therefore it follows that any rhythmic
increases in energy expenditure probably would be associated with
building and/or repair of webs. The lack of detection of a rhythm may
be an artifact in that the spiders were placed in the Warburg flasks
24 hr prior to measuring oxygen consumption. Webs were built during
this time and should have remained in good repair during the course
of the experiment.
I am tempted to speculate that the energy spent in the building
and maintenance of webs serves the same function, namely prey capture,
as that spent during the increased locomotory activity of the hunting
species. Certainly the web increases the effectiveness of its owner
in sampling the environment for prey, but there may be an added advan-
tage related to energy conservation in those spiders which do not make
a new web every day. Once the web is constructed its owner does not
have to raise his metabolic rate on a daily basis to insure capture
of prey, while hunting spiders must expend energy daily to emerge from
their retreats and move about in search of food. Unfortunately this
study does not provide data that would support this conjecture for
reasons discussed previously.
The importance of light as a Zeitqeber was demonstrated by the
correspondence between the standard metabolic rate of L. lenta with the
metabolic rate of an experimental group of this species which had been
kept in constant darkness. The results also suggest that even though
the conditions used to maintain the arachnids in the laboratory extin-
guished metabolic rhythms, measured rates of oxygen consumption would
be compatible with standard metabolic rates. The correspondence of
oxygen consumption values obtained under constant darkness with those
obtained from organisms when they are in an inactive phase under an
alternating light and dark cycle, if substantiated by future research,
suggests maintaining organisms under constant conditions to alleviate
the necessity of measuring metabolic functions at inconvenient times
in order to obtain standard metabolic rates.
To illustrate the interspecific variations in metabolic rates of
arachnids, the species metabolism-weight regression lines, and,where
appropriate, average metabolic rates versus weights were plotted in
Figure 6. In addition, the metabolic rates of all the arachnids were
combined and a pooled arachnid metabolism-weight line constructed in
Figure 6 as a reference line for the purpose of comparison.
With the exception of one point representing the average metabolic
rate of the solfugid, E. bilobatus, all of the values above the pooled
arachnid line represent spiders. They include L. lenta, P. reqius,
P. otiosus, A. tepidariorum, and T. sisyphoides. Immature stages of
P. reqius and L. lenta as well as adult T. rufipes have metabolic rates
which fall on the pooled arachnid line. The values below the pooled
arachnid line represent the metabolic rates of the amblypygid T.
marginemaculata; scorpions, C. hentzi and C. vittatus; three species
of theraphosid spiders; the trap-door spider, U. audouini; and the
L- O E
L C .J E
4) -J 4-' n
Q 0 0|
I 0:- u-'
( QI f
m vi -I .- .
0 &- -O
- *- I 4
L .n 3 *-
O -U --
C-- V1 *n .- N
3U -C: u
4a -- 1 (E
.C in 0 ol-
o -o U L
)-0.- .- IC
4-e C --- 4O T
U 4i 3 a|
4 C- + E4
0 t0 - 1 3
3 UO (0 0
T.*- I- 1 TO
C L .---
O30 vi Ti
4 4 *-' T
g .c -----I (o
Cl ^ ru 0
)-Q TO C O
+ + +
0 0 0 0
(J4/ O ) 31V1 3 O170 V[3H 901
cribellate spider F. hibernalis. The data on three species of scorpions
reported by Dresco-Derouet (1964) agree with the metabolic rate values
I obtained for scorpions.
Pooled regression lines were calculated for adult arachnids whose
metabolic rates are on or above the pooled regression line and for
those whose metabolic rates fall below the line in an attempt to
estimate the statistical significance of the difference in metabolic
rates of these groupings. Analysis of covariance, as described by
Snedecor (1956), was used for this purpose. Unfortunately the slopes
of the two lines were shown to be unequal thus preventing further
testing of the difference in elevation between the two lines. In a
partial attempt to circumvent this problem, the same test was conducted
on metabolic rate values of a pair of species, P. reqius and F.
hibernalis. The metabolism-weight regressions of these species fall
above and below the pooled arachnid line, respectively. The suitabil-
ity of using data obtained from these two species can be seen in
Figure 6: their weight ranges and slopes of their regression lines
coincide. With I and 156 degrees of freedom, the calculated f ratio
is 117.74 indicating that the difference between their respective
rates of oxygen consumption is statistically significant well beyond
the 1% level.
In considering these two groupings, i.e., arachnids with a high
or low metabolic rate as evidenced by the position of metabolic rate
values in respect to the pooled arachnid line, I came to the conclu-
sion that the various levels of the metabolic rate in arachnids are
associated with their evolutionary history, particularly in regards
to the fact that arachnids are predators. The arguments leading to
this conclusion are developed in the following paragraphs.
The composition of the group with a low metabolic rate includes
species that are representatives of primitive orders or are primitive
species within a more advanced arachnid order. I use the word primi-
tive in the sense that organisms so described are the most conservative
anatomically of their respective taxa. Whether an arachnid was con-
sidered primitive or not was made on the basis of information supplied
by Millot (1949), Stdrmer et al. (1955), and Lehtinen (1967). Assuming
that physiological conservatism would be correlated with anatomical
conservatism, I thought it valuable to discuss the evolution of
metabolic rates in arachnids. Since arachnids were one of the first
groups to move from an aquatic to a terrestrial habitat, they as
predators probably faced the problem of having an inconsistent food
supply. A low metabolic rate in this situation would certainly be of
value. Even in extant species, a low rate of energy expenditure would
be of selective advantage in that it would permit an animal to live in
an unstable environment with regard to prey density and/or an environ-
ment where the high rate of water loss concomitant with high metabolic
rates would be detrimental.
While a plausible argument may be constructed to explain the
adaptive significance of a low metabolic rate, the fact that certain
arachnids have relatively high metabolic rates deserves some attention.
Assuming that during the colonization of the terrestrial habitat prey
availability became more and more reliable, the metabolic rates of
arachnids could have increased beyond a required minimal level in
response to different selective pressures associated with increased
activity patterns, rapid rates of development, and increased reproduc-
tive potential. I am assuming that these physiological components
require a relatively high rate of energy expenditure and that maximum
metabolic rates are a function of standard metabolic rates in the sense
that the higher the standard metabolic rate, the higher the maximal
metabolic rate. With the exception of the solfugid, the arachnids
with relatively high metabolic rates are generally considered to have
undergone more evolutionary change than those arachnids with low
metabolic rates, at least in so far as the species used in this study
The association of a low metabolic rate with predatory habits and
relative lack of evolutionary change is not contradicted by reports
in the literature. Harvestmen are not solely dependent on living prey
as an energy source and represent one of the most advanced orders of
arachnids. A comparison of the metabolic rates obtained at 16 C for
two species of harvestmen, as reported by Phillipson (1962), with my
pooled arachnid line indicates that harvestmen have a rate of metabolism
three to four times higher even though my measurements were made at
20 C. I would argue that harvestmen evolved a relatively high metabolic
rate in response to some unknown selection pressure, and one of the
requisites of this high metabolic rate is a flexibility in terms of
accepting food sources in addition to living prey.
The fact that Dresco-Derouet (1967) reported a low metabolic rate
for the harvestman, Ischyropsalis luteipes Simon does not contradict
this line of reasoning since this species normally lives in caves.
As Frost (1959) points out, one of the main characteristics of a cave
environment is a scarcity of food. Derouet (1953) and Dresco-Derouet
(1960) also demonstrated that a low metabolic rate is associated with
a cavernicolous habitat in spiders.
Hemmingsen's (1960) study was selected as a basis for comparison
of arachnid metabolic rates with those of other polkilotherms because
of the comprehensive nature of the review. The values of the slopes
of the regression equations describing the relationship between weight
and metabolism of arachnids, as indicated in Table 5, do not differ
significantly from 0.75, a value that has been computed by Hemmingsen
(1960) from data obtained from a wide variety of organisms. To
detect any difference between levels of metabolism, the metabolic
rates of all the arachnids were pooled and then converted to their
caloric equivalents. A respiratory quotient of one was obtained from
experiments with P. reqius and was used to convert oxygen consumption
values to kilocalories. The comparison of the pooled arachnid line
and Hemmingsen's (1960) standard poikilotherm line is shown in Figure
7. The average metabolic rate of the arachnids sampled in this study
is about 40 to 50% lower than that of other poikilotherms over the in-
dicated weight range. I mentioned previously that certain spiders
had relatively high metabolic rates for arachnids, however, while
these rates approach Hemmingsen's (1960) standard poikilotherm line,
they still fall below it. Again I would argue that the low rate of
metabolism in the arachnids studied in comparison with other poikilo-
therms is the result of adaptation to a relatively inconsistent food
supply. Further remarks concerning the adaptive significance of high
and low metabolic rates will be made below where the results of the
effect of temperature on metabolic rate are discussed.
Although the preceding discussion is concerned with the adaptive
significance of the various levels of standard metabolism, no mention
had been made of the morphological factors which determine the various
0 31V0 0110OV 3W 00
rates of oxygen consumption for a given organism. Those arachnids
that have metabolic rates above the pooled arachnid line of Figure 6
carry out external respiration by means of book-lungs and/or trachea
whereas those arachnids with relatively low metabolic rates utilize
only book-lungs for this function. The implication that a tracheal
system is a factor associated with a high metabolic rate is supported
by certain evidence. The two individuals of T. marginemaculata, which
respire by means of book-lungs, had a metabolic rate of 41 and 23 ul
02/gm/hr while the two individuals of E. bilobatus, which use only
trachea, had measured rates of 246 and 249 ul 02/gm/hr. The spiders
that utilize both book-lungs and trachea have a higher metabolic rate
than those spiders that utilize only book-lungs. The latter group in-
cludes F. hibernalis, a spider having rudiments of a tracheal system.
It is doubtful, however, that this system functions in gas exchange in
this species. Even the spiders which have a relatively high metabolic
rate have rates inferior to those of insects as cited by Hemmingsen
(1960). The latter group utilize only trachea. The high metabolic
rates of the harvestmen also fit this scheme as they respire solely
by means of a tracheal system. This evidence suggests that a tracheal
system is more efficient in delivering oxygen to cells than is a book-
lung system associated with a circulatory system. However, considering
just spiders, the answer is not clear cut. Davies and Edney (1952)
demonstrated in the wolf spider, Lycosa amentata, that respiration
took place mainly through the book-lungs. Blocking of the spiracular
opening of the tracheal system with celloidin did not reduce the
oxygen consumption of these spiders. Dresco-Derouet (1960) repeated
this experiment on different spiders and obtained the same results. As
indicated in Table 6, P. regius has a well developed tracheal system
in addition to book-lungs, yet this spider has a weight specific
metabolic rate which closely approximates that of L. lenta, a spider
which has book-lungs and a poorly developed tracheal system. The
hypothesis advanced by Levi (1967) merits consideration at this point.
He suggests the tracheal system is probably important in reducing water
loss. As the rate of oxygen consumption increases there is a concomi-
tant increase in respiratory water loss, an intolerable consequence for
a small terrestrial organism if not corrected. The trachea may function
as a means of reducing this water loss because of a longer diffusion
distance and/or smaller spiracles relative to those associated with
book-lungs. The one spider with an extensive tracheal system, P.
regius, has its highest metabolic rate during the daytime when it would
be exposed to conditions conducive to high rates of water loss, i.e.,
relatively high temperatures and low relative humidities. On the
other hand, L. lenta, a spider with a poorly developed tracheal system,
is active at night when low temperatures and high relative humidities
would not be as conducive to high rates of water loss.
in summary, I suggest that a tracheal system is associated with
high metabolic rates because of its efficiency in transporting oxygen,
and perhaps more importantly because it reduces water loss. On the
basis of experiments of Davies and Edney (1952) and Dresco-Derouet (1960),
as well as my analysis of the association between metabolic rates and
tracheal systems of P. reqius and L. lenta, I concluded that in spiders
the tracheal system functions mainly during the active phases of the
daily cycle of the spider and not, at least to any great extent, when
the spider is quiescent.
If the latter assumption is true, then analysis of the respiratory
surface area of only the book-lungs is justified from the standpoint
of associating standard metabolic rates with respiratory surface areas.
Since estimation of the surface area of a book-lung requires an exces-
sive amount of time, the analysis of the effect of this measure on
metabolic rates was conducted interspecifically rather than intraspe-
cifically. Table 9 summarizes the data used in this analysis while
Figure 8 shows a graph of metabolism versus respiratory surface area.
AVERAGE WEIGHT, METABOLIC RATE AND
BOOK-LUNG SURFACE AREA IN ADULT SPIDERS
Species Avg Wt Avg Metabolic Rate Avg Book-Lung Surface
(mg) (ul 02/hr) area
L. lenta 970 90 333
P. reqius 568 54 258
F. hibernalis 571 29 142
A. tepidariorum 73 25 39
T. sisyphoides 110 17 49
T. rufipes 25 5 14
The results listed in Table 9 indicate a general correspondence
between metabolic rate and respiratory surface area, i.e., the larger
the surface area the higher the rate of metabolism. However, the
weight component complicates the analysis as there are only two species,
F. hibernalis and P. reqius, which have roughly the same average weight.
Figure 8. Relationship between book-lung surface area and oxygen
consumption in spiders. (a) L. lenta; (b) P. reqius;
(c) F. hibernalis; (d) T. sisyphoides; (e) A. tepidariorum;
and (f) T. rufipes.
10P0 S E ( 02
LOG RESPIRATORY SURFACE AREA (mm2)
Regression lines describing the relationships of metabolism and surface
area to weight were calculated using the average values listed in
Table 9 to provide standards in an attempt to correlate deviations in
the metabolism-weight relationship with deviations in the respiratory
surface area-weight relationship. The logarithemic form of these equa-
log metabolism = 0.6320 (log wt) 0.0408 and
log surface area = 0.9398 (log wt) 0.0127
The sample correlation coefficients for these relationships are 0.91
and 0.99, respectively. These equations were used to calculate ex-
pected metabolic rates and surface areas at the average weights for
each species. The calculated values were then compared with the ob-
served values to see whether the deviations between calculated and
observed values of metabolic rate corresponded to deviations between
calculated and observed values of surface area. The results of these
procedures are shown in Figure 9. In every case, where there is a
metabolic rate higher than predicted on the basis of the metabolism-
weight regression equation, there is a larger than predicted surface
area. The converse situation also holds. This analysis suggests that
the rate of oxygen consumption is a function of the respiratory surface
area of the book-lungs. To ascertain the degree of statistical signifi-
cance of the relationship between metabolic rate and respiratory surface
area, a sample partial correlation was computed. This method of analy-
sis eliminates the effect of weight. The calculated sample partial
correlation is 0.86 which, with 3 degrees of freedom, is significant
at the 10% level but not at the 5% level. The hypothesis that the
greater the amount of surface area the higher the metabolic rate does
not imply that other factors such as cardiac output and properties
Figure 9. Deviations of actual versus calculated values of oxygen
consumption and respiratory surface area. Abscissa, surface
area deviations. Ordinate, oxygen consumption deviations.
(a) A. tepidariorum; (b) L. lenta; (c) P. reqius; (d) T.
sisyphoides; (e) T. rufipes; and (f) F. hibernalis.
of respiratory pigments may not be important in regulating the level
of oxygen consumption. The former is certainly of some importance
since Sherman and Pax (1968) discovered marked variations in heart rate
corresponding to levels of locomotory activity in the spider Geolycosa
The hypothesis that disproportionate increases in supportive
tissues that occur as animals get larger is responsible for the rate of
increase in metabolic rate with an increase in weight is not validated
by the results of this study. Regression equations describing the
relationships of exoskeleton weight and metabolically active tissue
weight to total weight were calculated, and are reported in Table 10.
In the case of the three species investigated, the weight of metaboli-
cally active tissue increases in direct proportion to total weight,
i.e., the slope of the equation describing this relationship approxi-
mates a value of one. If the hypothesis were to hold, the slopes of
the regressions of weights of netabolically active tissue and rates of
metabolism to total weight should equal one another. Comparisons of
these values reported in Tables 5 and 10 indicate this is not the case.
Unfortunately, the lack of information pertaining to the biochemical
composition of spiders precludes testing the assumption that the
exoskeleton constitutes the bulk of metabolically inactive tissue.
The apparent anomaly where exoskeleton weight increases disproportion-
ally with an increase in total weight while active tissue weight in-
creases in direct proportion, at least in the case of L. lenta, is
resolved when these relationships are analyzed. The form of the equa-
tions are determined by two variables, namely the slope and the inter-
cept. While exoskeleton weight may increase disproportionally with an
L co co0 c
n L (
o 6 C
S L. t
0o 0) C 0)
In .i-J *-
LE 0-> c4
0 (0 -
) .- -
-0 I (
u- 0 4-
0- .- >
E-- E 0
U (0 (
0 0 0-)
E t-- -
O0D 0 -
l (- J- ( 0
CL D C 0
-C o 0-
C0 r- O'N
increase in total weight, the mathematical relationship between active
tissue weight and total weight can be one of direct proportion due to
the higher intercept in the latter. These relationships are indicated
in Table 10.
The results pertaining to the effect of temperature on metabolic
rate, as recorded in Table 8, indicate oxygen consumption in spiders
is not independent of rapid changes in temperature. The calculated
average Qi0 values for temperature ranges of 10 to 20 C and 20 to 30 C
are 3.3 and 2.4, respectively. These values are in contrast with the
findings of Newell (1966). He claimed the metabolic rates of poikilo-
therms are largely independent of rapid temperature fluctuations with
resulting Q10 values ranging from 1.2 to 1.3 provided that metabolic
rates of quiescent organisms were measured at frequent intervals. My
experimental procedures agree with these requirements in that oxygen
consumption was measured at 10 to 15 minute intervals after a rapid
10 C temperature change at a time of day when the spiders were inactive.
The correspondence of my results with those of Tribe and Bowler (1968)
suggests temperature independence in poikilotherms is not as wide-
spread as Newell (1966) implies.
The results recorded in Table 8 indicate previous thermal history
does have an effect on metabolic rate when measured at a new temperature.
In every case, except F. hibernalis, acclimation to a low temperature
of 10 C resulted in a significantly higher metabolic rate at 30 C than
would be produced when the spiders were acclimated at 20 C or 30 C.
This situation is depicted graphically in Figure 3 where the metabolic
rate-temperature curves show a rotation around 20 C. Tiis phenomenon
appears to be associated with the ability of individual species to
compensate at high temperatures by lowering their metabolic rates
after exposure to the new high temperatures. For example, individuals
of P. regius, when acclimated at 20 C for three weeks, have an initial
average metabolic rate of 237 ul 02/gm/hr at 30 C. After a six-day
exposure to this temperature, the average metabolic rate decreased to
151 ul 02/gm/hr. In each case where compensations to a high temperature
were detected, the degree of compensation was not complete because the
rate of oxygen compensation did not decrease to the level recorded at
20 C. The magnitude of decrease in metabolic rate corresponds to what
Precht (1958) terms a 'type 3' compensation. The degree of compensa-
tion, computed from the average metabolic rates initially obtained at
30 C after acclimation for three weeks at 20 C and those obtained at
30 C after a six-day exposure to this temperature, involve reductions
of 20%o for L. lenta, 36% for P. regius, and 31% for T. rufipes. In all
cases, the maximal compensation was reached by the third day of exposure
to 30 C.
In trying to relate the findings of the temperature studies to
standard metabolic rates, I noticed the spiders which showed a decrease
in oxygen consumption during exposure to 30 C all had relatively high
metabolic rates at 20 C as indicated in Figure 6. The one spider which
did not demonstrate compensation at 30 C had a relatively low metabolic
rate at 20 C. This correlation suggests homr'ostatic potential in
regards to temperature is perhaps limited by the relative rate of metab-
olism. Making the assumption there is a certain minimum level of energy
expenditure required to keep arachnid protoplasm in a healthy statc,
it seems reasonable to have a standard metabolic rate at this relatively
low level under certain conditions. As mentioned previously, I suspect
the movement from an aquatic to a terrestrial habitat that arachnids
made in the geological past presented the problem of a limited food
supply. However, as time passed and more and more organisms colonized
the terrestrial habitat, the luxury of a high metabolic rate was not
selected against, on the contrary it probably possessed certain adaptive
adjuncts. One of these adjuncts may relate to'the relatively large
temperature fluctuations inherent in a terrestrial environment.
Consider a spider that has a low metabolic rate at an intermediate
temperature. This animal is living close to its minimal level of re-
quired energy expenditure. A rise in environmental temperature of 10 C,
through its effects on the kinetics of enzyme systems, would result in
roughly two fold increases in both minimal and standard metabolic
rates. Since the new minimal and standard metabolic rate approximate
one another in this case, any reduction in energy expenditure, which
might well be desirable from the standpoint of the limitation of energy
resources, would be incompatible with maintaining life. On the other
hand, a spider with a relatively high metabolic rate at the intermediate
temperature, that is to say its standard metabolic rate is appreciably
above minimal requirements, when exposed to a higher temperature would
also increase its required minimal and standard metabolic rates. How-
ever, since this new rate would be above minimal requirements, a
decrease in oxygen consumption could take place. This reduction in
energy expenditure would be advantageous when the amounts of food re-
quired to support the different levels of metabolism are considered.
This model is depicted graphically in Figure 10 using the data obtained
from F. hibernalis and P. reqius. I must admit that this explanation
is speculative. The major difficulty stems from the acceptability of
Figure 10. Relationship between temperature and standard metabolism
in P. regius and F. hibernalis. (a) minimal required energy
expenditure at 20 C; (b) minimal required energy expenditure
at 30 C; (c) compensation at 30 C in P. reqius.
c P. reqius
20 C o c
the assumption that there is a certain minimal level of energy expen-
diture which may be different in certain cases from that defined as
the standard metabolic rate.
The lack of compensation to the 10 C temperature is not surprising.
As predators of insects it would be of little value for spiders to
increase their metabolic rate at low temperatures since the low temper-
atures would also reduce the activity of their prey. I would even
suggest that the inverse compensation exhibited by F. hibernalis, a
web builder, reflects an adaptation to the greater inhibitory effect
of low temperature on flying versus walking insects.
Anderson, J. F. 1966. The excreta of spiders. Comp. Biochem. Physiol.
Benedict, F. G. 1938. Vital energetic: A study in comparative basal
metabolism. Carnegie Inst. of Washington Publ. 503, 215 p.
Berthet, P. 1963. Mesure de la consommation d'oxygene des Oribatides
(Acariens) de la litiere des forts. p. 18-31. In J. Doeksen and
J. Van Der Drift, (ed.), Soil organisms. North-Holland Publ. Co.,
Brown, M. E. 1957. The physiology of fishes, Volume I. Metabolism.
Academic Press Inc., New York. 447 p.
Bullock, T. H. 1955. Compensation for temperature in metabolism and
activity of poikilotherms. Biol. Rev. 30:311-342.
Cloudsley-Thompson, J. L. 1960. Adaptive functions of circadian
rhythms. Cold Spring Harbor Symposia on Quantitative Biol.
Davies, M. E. and E. B. Edney. 1952. The evaporation of water from
spiders. J. Exptl. Biol. 29:571-582.
Derouet, L. 1953. M6tabolisme compare de deux Araignees, I'une
troglophile, I'autre epig6e obscuricole. influence de variations
brusques de temperature et d'humidite. Premier Congr. Intern.
Speleologie Publ. Paris. 3:237-240.
Dresco-Derouet, L. 1960. Etude biologique compare de quelques
especes d'Araign6es lucicoles et troglophiles. Arch. Zool. Exptl.
1961. Le m6tabolisme respiratoire des Scorpions. I.
Existence d'un rythme nycth6meral de la consummation d'oxygene.
Bull. Museum Natl. Hist. Nat. 32:553-557.
_1964. Le metabolisnie respiratoire des Scorpions. II.
Mesures de 1'intensite respiratoire chez quelques especes a
diff6rentes temperatures. Bull. Museum Natl. Hist. Nat. 36:97-99.
1967. Biologic et metabolisme respiratoire d'Ischvropsalis
luteipes Simon (Opiliones) adulte, au laboratoire. Ann.
Frost, S. W. 1959. Insect life and insect natural history. 2nd ed.,
Dover Publ. Inc., New York. 526 p.
Gertsch, W. J. 1949. American spiders. D. Van Nostrand Co., Inc.,
Princeton. 285 p.
Hemmingsen, A. M. 1960. Energy metabolism as related to body size
and respiratory surfaces and its evolution. Rept. Steno Mam.
Hospital and Nordisk Insulinlaboratorium, Part II, 9:1-110.
Hoar, W. S. 1966. General and comparative physiology. Prentice-Hall,
Inc., Englewood Cliffs. 815 p.
Lehtinen, P. T. 1967. Classification of the Cribellate spiders and
some allied families, with notes on the evolution of the suborder
Araneomorpha. Ann. Zool. Fennici 4:199-468.
Levi, H. W. 1967. Adaptations of respiratory systems of spiders.
Millot, J. 1949. Trait6 de zoologie. Vol. 4, Onychophores,
Tardigrades, Arthropodes, Trilobitomorphes, Chelicerates. P-P.
Grass (ed.). Masson et Cie. Paris. 979 p.
Mitchell, R. 1964. The anatomy of an adult chigger mite Blankaartia
acuscutellaris (Walch). J. Morphol. 114:373-391.
Newell, R. C. 1966. Effect of. temperature on the metabolism of
poikilotherms. Nature 212:426-428.
and H. R. Northcroft. 1967. A re-interpretation of the
effect of temperature on the metabolism of certain marine
invertebrates. J. Zool. 151:277-298.
Petrunkevitch, A. 1933. An inquiry into the natural classification
of spiders based on a study of their internal anatomy. Trans.
Connecticut Acad. Sci. 31:299-389.
Phanuel, G. J. 1967. Orientation and biorhythms in jumping spiders
(Araneida: Salticidae). Ph.D. Dissertation, Univ. Florida. 79 p.
Phillipson, J. 1962. Respirometry and the study of energy turnover
in natural systems with particular reference to harvest-spiders
(Phalangiida). Oikos 13:311-322.
Precht, H. 1958. Patterns of temperature adaptation, p. 50-78.
In C. L. Prosser, (ed.), Physiological adaptation, American
Physio. Soc., Washington, D. C.
Prosser, C. L. and F. A. Brown. 1961. Comparative animal physiology.
2nd ed., W. B. Saunders Co., Philadelphia. 688 p.
Sherman, R. G. and R. A. Pax. 1968. The heartbeat of the spider,
Geolycosa missouriensis. Comp, Biochem. Physiol. 26:529-536.
Snedecor, G. W. 1956. Statistical methods applied to experiments in
agriculture and biology. 5th ed., Iowa State Univ. Press, Ames.
Stbrmer, L. A. Petrunkevitch, and J. W. Hedgpeth. 1955. Treatise on
invertebrate paleontology. R. C. Moore (ed.), Part P, Arthropoda
2, Chelicerata with sections on Pycnogonida and Palaeoisopus.
Geol. Soc. America and Univ. Kansas Press, Lawrence. 181 p.
Tenney, S. M. and J. E. Remmers. 1963. Comparative quantitative
morphology of the mammalian lung: diffusing area. Nature 197:54-56.
Tribe, M. A. and K. Bowler. 1968. Temperature dependence of "standard
metabolic rate" in a poikilotherm. Comp. Biochem. Physiol.
Whitford, W. G. and V. H. Hutchison. 1967. Body size and metabolic
rate in salamanders. Physiol. Zool. 2:127-133.
John Francis Anderson was born July 25, 1936, at Hartford,
Connecticut. He graduated from Bulkeley High School in June, 1954.
From 1954 until 1958, Mr. Anderson served in the United States Marine
Corps. Following his release from active duty he entered Central
Connecticut State College at New Britain, Connecticut, where he re-
ceived his Bachelor of Science degree with honors in June, 1962.
In September, 1962, he entered the Zoology Department at the University
of Florida where he received his Master of Science degree in April,
1965. He continued graduate studies toward the degree of Doctor of
Philosophy at the University of Florida while serving as a research
and teaching assistant.
Mr. Anderson is married to the former Marjorie Ann Barrett. He
is a member of the American Society of Zoologists and Sigma Xi.
This dissertation was prepared under the direction of the chair-
man of the candidate's supervisory committee and has been approved by
all members of that committee. It was submitted to the Dean of the
College of Arts and Sciences and to the Graduate Council, and was
approved as partial fulfillment of the requirements for the degree of
Doctor of Philosophy.
Dean, Graduate School
?, n. ~e