1137
The Journal of Experimental Biology 198, 1137–1149 (1995)
Printed in Great Britain © The Company of Biologists Limited 1995
REGULATORY RESPONSES OF THE COXAL ORGANS AND THE ANAL
EXCRETORY SYSTEM TO DEHYDRATION AND FEEDING IN THE SPIDER
PORRHOTHELE ANTIPODIANA (MYGALOMORPHA: DIPLURIDAE)
A. G. BUTT* AND H. H. TAYLOR†
Department of Zoology, University of Canterbury, Private Bag 4800, Christchurch, New Zealand
Accepted 3 January 1995
Summary
The roles of coxal and anal excretion in the regulation of
haemolymph osmolality, [Na+] and [K+] were studied in the
mygalomorph spider Porrhothele antipodiana (mass
0.7–1.0 g) under differing conditions of feeding and
hydration state. Dehydration of starved spiders by removal
of drinking water caused progressive mass loss at a rate of
about 2.5 % of initial body mass per day and was associated
with increases in the whole-body [Na+] and [K+] and in the
osmolality, [Na+] and [K+] of the haemolymph. On
provision of prey, feeding partially restored this mass loss
but further elevated body and haemolymph ion
concentrations. Dehydration reduced fluid excretion by the
anal excretory system and the four coxal organs in both
starved and feeding spiders. Starved hydrated spiders
initially produced anal urine at 5 ml day21 and this was
progressively reduced to zero after 4 days of water
deprivation. Spiders dehydrated to less than 12 % mass loss
would nevertheless feed and this initiated a small postprandial anal diuresis (<5 ml day21 compared with
>30 ml day21 in fed hydrated spiders). Coxal fluid was
produced by dehydrated spiders only during feeding and
was delivered into the prey, the rate of production by single
organs decreasing from about 19 ml h21 g21 body mass in
hydrated spiders to about 4 ml h21 g21 body mass in spiders
dehydrated to 11 % mass loss.
There was an increase in urine [K+] and in the rate of
anal K+ excretion associated with ad libitum feeding in
dehydrated spiders. However, urine [Na+] and the rate of
anal excretion of Na+ were not increased by feeding. This
was associated with an increase in [K+] of the stercoral fluid
above that observed in either fed or starved hydrated
spiders, but no significant change in [Na+]. Conversely,
[Na+] of the coxal fluid produced during feeding was
increased by dehydration whereas [K+] was not. These
observations are consistent with the previously postulated
roles of the coxal organs (Na+) and anal system (K+) in the
excretion of ions ingested with the prey.
Full elimination of the prey ions was accomplished only
after drinking water was resupplied, which initiated
further anal and coxal diureses. Smaller anal and coxal
diureses also occurred on rehydration of unfed spiders. The
production of coxal fluid in the absence of prey is further
evidence that the coxal organs have a true excretory
function besides, presumably, assisting ingestion. During
dehydration and feeding, P. antipodiana, unlike many
insects, is unable simultaneously to conserve water and to
eliminate ions by production of a highly concentrated
excretory fluid. Both coxal fluid and anal urine were
approximately iso-osmotic to the haemolymph and the
urine was markedly hypo-ionic.
Key words: Arachnida, spider, excretion, dehydration, water balance,
osmoregulation, ionic regulation, feeding, coxal organs, Porrhothele
antipodiana.
Introduction
Despite their large surface-to-volume ratio and consequent
susceptibility to desiccation, arthropods, and in particular
insects, and spiders and other arachnids, are familiar
inhabitants of virtually all terrestrial habitats. The success of
the insects is attributed to a range of mechanisms that limit
water loss. These include behavioural adaptations, a highly
impermeable cuticle (Wigglesworth, 1945; Ebeling, 1974;
Gilby, 1980; Hadley, 1994a), control of the respiratory
openings (Bursell, 1957; Miller, 1964a,b; Loveridge, 1968;
Krasfur, 1971a,b; Hadley, 1984b) and production of highly
concentrated urine (Wigglesworth, 1931; Ramsay, 1952, 1955,
1964; Phillips, 1964, 1970, 1981; Maddrell, 1981). Spiders
employ a similar range of devices that limit evaporative losses
across the body and respiratory surfaces (Davies and Edney,
1952; Cloudsley-Thompson, 1957; Stewart and Martin, 1970;
Seymour and Vinegar, 1973; Humphreys, 1975; Hadley, 1978,
*Present address: Department of Physiology, Medical School, University of Otago, PO Box 913, Dunedin, New Zealand.
†Author for correspondence.
1138 A. G. BUTT
AND
H. H. TAYLOR
Robinson and Paim, 1969; Hadley and Quinlan, 1989; Paul and
Fincke, 1989). However, the responses of their excretory
systems to the demands of the terrestrial environment are
poorly understood.
Two excretory systems are present in the spiders. The basic
arrangement in the mygalomorph Porrhothele antipodiana was
described by Butt and Taylor (1986, 1991). The first, referred
to as the anal system, resembles the excretory system of
insects. It consists of two systems of branching tubules opening
into the gut (Malpighian tubules and midgut diverticula) and a
cuticle-lined posterior expansion similar to the insect rectum
(termed the stercoral pocket). The second system consists of
the coxal organs, which, in P. antipodiana, open at the base of
the first and third legs and discharge fluid into a cuticular
groove leading to the oral region. The coxal organs are derived
from coelomoducts and are structurally similar to filtrationtype excretory organs in other arthropods (Goodrich, 1945;
Clarke, 1979). In most spiders, the relative importance of these
two systems is unknown. The trend towards reduction of the
coxal organs in the more advanced Araneae has led to
suggestions that coxal organs no longer serve an excretory
function and are obsolete (Buxton, 1913; Millot, 1949).
However, for the primitive mygalomorph Porrhothele
antipodiana, evidence has been presented that both the coxal
organs and the anal system contribute to ion excretion
associated with feeding (Butt and Taylor, 1986, 1991). In
hydrated spiders feeding on prey of normal salt content, the
activity of the coxal organs is limited to the period of feeding.
Coxal fluid is directed over the prey and much of it is
reingested, presumably facilitating ingestion. However, a
portion of the Na+-rich coxal fluid is left in the discarded prey
debris and effectively excreted. Furthermore, the coxal organs
show regulatory responses to changes in the Na+ balance of the
spiders. Increasing the Na+ load, either by direct injection into
the haemolymph or by salt-loading the prey, elevates both the
[Na+] and volume of coxal fluid and, in these circumstances,
coxal excretion occurs not only during feeding but also at other
times. A more limited response of the coxal organs is seen with
dietary K+ loading.
Unlike the coxal excretion, anal excretion occurs after the
meal. A post-prandial diuresis occurs over several days which,
together with the coxal excretion, eliminates most of the ions
ingested. The coxal organs are primarily responsible for Na+
excretion and the anal system excretes mainly the K+ ingested
with the meal (Butt and Taylor, 1986). The anal system is,
however, capable of excreting large quantities of Na+ and
responds to both dietary Na+ and K+ loading with a more
prolonged diuresis and increased concentrations of the
appropriate ion (Butt and Taylor, 1991). Intermittent
production of anal urine is maintained indefinitely in hydrated
spiders and it is also a vehicle for nitrogenous excretion.
Nitrogenous excretory products are absent from the coxal fluid
(Butt, 1983).
Salt loading of the prey and saline injection are somewhat
artificial physiological challenges. In contrast, a lack of fresh
drinking water for several days must be common for P.
antipodiana in its terrestrial/supralittoral habitat. Butt and
Taylor (1986) observed that hydrated P. antipodiana produced
excretory fluids hypo-ionic to their haemolymph and
concluded that the spider must balance evaporative water
losses, and also support the elimination of ingested prey ions,
by imbibing hypo-ionic fluids. As selective foraging for prey
with hypo-ionic body fluids is not practicable for an
opportunistic predator such as P. antipodiana, it appeared that
feeding must be accompanied by drinking fresh water. Thus,
it is of interest to determine whether and how P. antipodiana
is able to deal with prey during short-term water shortages.
Such information may also provide further insight into the
putative excretory functions of the two systems. In this paper,
we examine the functioning of the coxal organs and the anal
excretory system in relation to dehydration and to feeding
combined with water deprivation. The following questions
were of interest. Will water-deprived or dehydrated spiders
capture prey? Do dehydrated spiders partition the excretion of
sodium and potassium between the coxal organs and the anal
system in a pattern similar to that of hydrated spiders? Do
changes in either the volume or composition of the coxal fluid
and anal urine support their proposed roles in the excretory
regulation of ions? Is either system capable of producing a
highly concentrated excretory fluid that would allow P.
antipodiana to maintain a positive water balance through
feeding alone, despite evaporative water losses?
Materials and methods
Porrhothele antipodiana (Walckenaer) (0.7–1.0 g, unsexed)
were collected from Whalers Bay, Kaikoura Peninsula, New
Zealand. Maintenance of animals in the laboratory and
experimental techniques have been described previously (Butt
and Taylor, 1986, 1991) but are briefly summarised here. All
experiments were carried out at room humidity (60–80 % RH)
and temperature (20±2 ˚C). Prior to experiments, spiders were
hydrated by provision of drinking water ad libitum but starved
for a period of 3 weeks. During the experiments, spiders were
kept in individual clean glass chambers lined with Whatman
542 filter paper. The filter paper lining was changed at varying
intervals depending upon the rate of excretion. When required,
water was provided by placing absorbent cotton wool rolls
soaked in tap water in small troughs on the floor of the
containers. The food provided was live cockroach nymphs
taken from a culture maintained on a constant diet.
Rates of anal urine production and Na+ and K+ excretion
were determined as described previously (Butt and Taylor,
1986). The excretion of coxal fluid at times other than during
feeding was monitored by labelling the spiders’ haemolymph
with 22Na (Butt and Taylor, 1991). Regions of high 22Na
activity on the filter paper indicated areas of coxal excretion.
After removal of regions of anal excreta (visualized by
fluorescence), these coxal fluid residues were isolated by
subdividing the filter paper into small pieces and measuring the
22Na activity of the individual pieces. Areas of dry coxal fluid
were transferred to appropriate volumes of 1 % HNO3 and their
Spider excretion 1139
Na+ and K+ contents determined by atomic absorption or
emission spectroscopy. The quantities of ions and water
ingested from the prey were calculated from the change in dry
mass of prey/debris, assuming proportional ingestion of all
constituents (Butt and Taylor, 1986), compared with the values
obtained by analysis of a group of similar-sized cockroach
nymphs taken from the same culture at the same time (typical
composition:
Na+,
0.162 mmol g21 dry mass;
K+,
21
21
0.349 mmol g dry mass;
water,
3.0 ml g dry mass).
Quantities of Na+ and K+ excreted in coxal fluid into the prey
debris were calculated from the difference between the
calculated quantities ingested and the measured change in prey
ion content (P. antipodiana finely macerates and partially
digests the prey externally so that it is unlikely that selective
feeding on haemolymph or cells could lead to differential ion
uptake; see Butt and Taylor, 1986, for further discussion of this
point). Volumes of coxal fluid excreted into the prey or onto
the paper were calculated from the quantities of Na+ deposited
and from measurements of the [Na+] of coxal fluid collected
from similarly dehydrated spiders.
Coxal fluid was collected from single anterior openings, as
reported by Butt and Taylor (1991). Briefly, the spider was
restrained ventral side up and the groove leading from the
coxal gland to the mouth was blocked with dental wax. The
spiders were then provided with freshly killed prey and
encouraged to eat by stimulating the chelicerae with the prey.
Coxal fluid was sucked into calibrated silicone-coated glass
capillaries and, after volume estimation, it was stored briefly
under liquid paraffin before analysis. Methods for collection of
haemolymph, anal urine and stercoral fluid and measurements
of their volumes, osmolalities and ionic compositions were as
reported by Butt and Taylor (1986). Note that anal urine refers
to fluid voided from the anus and, with faecal material, forms
the excreta. Stercoral fluid was collected directly from the
stercoral pocket and presumably represents incompletely
processed urine.
Means are given ± one standard error in the text, tables and
figures. Where error bars are not shown, they are smaller than
the symbols. Statistically significant differences between
means were determined using Student’s t-test for two means
and using one-way analysis of variance (ANOVA) and the
Tukey–Kramer test for multiple comparisons.
Protocols
Specific details of the four experimental series reported in
the Results are summarised here.
Series I
Twenty spiders were divided into two groups of similar
mean mass. The first group was provided with live cockroach
nymphs ad libitum without water for 6 days, while the second
group was starved without water for 6 days. Total masses, rates
of urine production and rates of cation excretion were
determined daily and, at the end of a 6 day period, haemolymph
samples and the contents of the stercoral pocket were collected
from each spider. The water and cation contents of the body
were then determined.
Series II
Six groups of ten spiders of similar mean mass were used.
One group was provided with a single meal of three live
cockroach nymphs and free access to water. A second group
was provided with a similar meal without water. The remaining
groups were dehydrated for 12, 48, 72 or 96 h before they were
also given a meal without water. The volume of urine produced
by those spiders that actually captured and fed on the
cockroaches was measured for 3 days during and after feeding.
Series III
Thirty-two spiders were divided into two groups of similar
mean mass. All were injected with 22Na to aid in the location
of coxal fluid outside the prey remains. After 48 h of recovery,
the spiders were placed in individual containers lined with
Whatman 542 filter paper without water at room temperature
and humidity for 3 days. The 16 spiders of the experimental
group were then provided with three cockroach nymphs for 1
day. They were dehydrated for a further 3 days and 10 spiders
were taken from the group for analysis of body ion content.
The remaining six spiders in this group were provided with
water for 3 days to allow them to rehydrate before analysis of
body ions. The changes in body mass, urine production, rates
of anal Na+ and K+ excretion and rates of coxal Na+ and K+
excretion were measured daily. The prey debris was collected
and analysed for estimation of quantities of water and ions
ingested and excreted in coxal fluid. The control group of
spiders was treated identically but was not provided with a
meal.
Series IV
Spiders were dehydrated for varying periods at room
temperature and humidity by depriving them of water. Mass
loss was measured and coxal fluid was collected directly from
single anterior coxal openings from the spiders immobilised
ventral side uppermost during feeding (Butt and Taylor, 1991).
The rate of fluid production was estimated and haemolymph
samples were collected. The [Na+] and [K+] and osmotic
pressure of the fluid samples were then measured.
Results
Changes in pools of water and ions and dry mass in fed and
unfed spiders during 6 days of dehydration (series I)
The ability of P. antipodiana to maintain its fluid and
electrolyte balance through feeding alone was investigated by
comparing the salt and water content, haemolymph
composition and stercoral fluid composition of spiders either
starved and dehydrated or fed and dehydrated for 6 days.
Mass changes, Na+ and K+ contents and concentrations
Starved spiders, deprived of water, lost 16.4 % of their body
H. H. TAYLOR
Table 1. Comparison of the live mass, dry mass and the
water and cation contents of spiders deprived of water for
6 days and either starved or fed ad libitum
Starved
Initial total mass, ITM (g)
(g g–1 ITM)
Anal urine production
(ml day−1)
6
4
2
0
Final body
Total (mmol)
Overall concentration
(mmol kg–1 FTM)
41±2
67±3
61±3***
87±4***
Final body Na+
Total (mmol)
Overall concentration
(mmol kg–1 FTM)
52±2
85±4
70±3***
99±3**
K+
ITM, initial total live mass; FTM, final total live mass.
Series I spiders. Pretreatment: 3 weeks starvation with water
supplied. Values are means ±1 S.E.M., N = 10 in all cases.
Significance levels of differences between mean values of starved
and fed spiders: *P<0.05; **P<0.01; ***P<0.001 (Student’s t-test).
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
Time (days)
4
5
6
0.30 B
0.20
0.10
0
0
1.5
C
1.0
0.5
0
0
0.736±0.053 0.703±0.013
0.998±0.015***
0.673±0.008**
0.327±0.012***
0.674±0.008*
A
8
0
Fed
0.836±0.013
0.592±0.020
0.257±0.005
0.703±0.009
Final relative total mass
Final water content (g g–1 ITM)
Final relative dry mass (g g–1 ITM)
Final water fraction (g g–1 FTM)
water content as a fraction of final mass was actually
significantly
lower
in
fed,
dehydrated
spiders
(0.674 g g21 final total mass) than in unfed, dehydrated spiders
(0.703 g g21 final total mass). Both values are lower than the
calculated initial value of 0.752 (above) and also lower than
that for a similar group of unfed, hydrated spiders
(0.748±0.007; Butt and Taylor, 1986).
Feeding increased the cation content of the spiders. After the
6 days of dehydration, the total body Na+ was 70 mmol in the
fed spiders and 52 mmol in the unfed spiders. Total body K+
Anal Na+ excretion
(mmol day−1)
mass (Table 1) at a steady rate over 6 days. Assuming this
represented only water loss, it may be calculated that the initial
water content of the spiders was 75.2 % of total mass and that
21.8 % of their initial body water was lost (there was negligible
faecal production; based on the final water and dry matter
values given in Table 1). The spiders provided with
cockroaches ad libitum maintained their mass during this
period (Table 1). In fact, an initial mass gain of about 4 % was
observed in this group, peaking on the third day. Initially, the
feeding behaviour was similar to that seen in hydrated spiders.
At least seven of the ten spiders captured prey and fed on each
of the first 3 days, leaving discarded prey debris as grey paste
in which no specific parts were identifiable (Butt and Taylor,
1986). As dehydration progressed, fewer spiders captured prey
and those that did often did not feed or only partially consumed
the prey. On the fourth day, only three spiders actually fed on
killed prey and, on the final day, only one spider fed. In
contrast, if provided with food plus water, 80 % of the spiders
fed daily (data not shown).
The very low rates of urine production (Fig. 1A) and
negligible production of faeces suggest that the fall in body
mass of the dehydrated starved spiders was due primarily to
evaporation of water. Relative to their initial body masses, fed
spiders had a significantly higher final water content than
starved spiders (0.673 and 0.592 g g21 initial total mass,
respectively), presumably as a result of ingestion of fluid
from the prey. These data indicate that at least
81 mg g21 initial total mass of water was obtained from the prey
(more if feeding spiders experienced higher evaporative losses
than unfed ones). The fed spiders also gained dry mass from
the prey (70 mg g21 initial total mass; Table 1). Consequently,
Anal K+ excretion
(mmol day−1)
AND
D
2.0
Molar Na+/K+ ratio
of excreta
1140 A. G. BUTT
1.5
1.0
0.5
0
0
Fig. 1. The effect of feeding Porrhothele antipodiana with cockroach
nymphs ad libitum without free water on the daily anal excretion of
(A) urine, (B) Na+ and (C) K+. (D) Molar Na+/K+ ratio of excreta.
Open symbols, fed spiders; filled symbols, starved controls. Points are
means ± S.E.M.; N=10 for both groups. Series I spiders; mean initial
mass, 0.70 g (fed) and 0.74 g (unfed). Simultaneous points are
displaced on the abscissa for clarity.
Spider excretion 1141
was similarly elevated in the fed spiders (61 mmol) compared
with unfed spiders (41 mmol). Despite gains of water and solids
from the prey, the overall body concentrations of these ions
were also higher in the fed group ([Na+], 99 and [K+],
87 mmol kg21 wet tissue) than in the starved spiders ([Na+], 85
and [K+], 67 mmol kg21 wet tissue).
accumulated similar quantities (about 20 mmol) of Na+ and K+
in the body during feeding (Table 1), implying an additional
route for the excretion of Na+, as shown below.
Composition of haemolymph and stercoral fluid
The compositions of the haemolymph and stercoral fluid of
the fed and the unfed dehydrated spiders are given in Table 2,
where they are compared with measurements from hydrated
spiders in a parallel experiment reported previously (Butt and
Taylor, 1986). Dehydration of starved spiders significantly
increased the osmolality of the haemolymph from 436 to
579 mosmol kg21, an increase largely accounted for by the
increase in the [Na+] of the haemolymph from 196 to
297 mmol l21 (assuming a similar increase in anions). The
increase in osmolality suggests the spiders did not
osmoregulate, being of the order predicted from the water loss
(i.e. 558 mosmol kg21 based on 21.8 % water loss given
above). The final [Na+] is rather higher than the value expected
from passive haemoconcentration (251 mmol l21).
Feeding of dehydrated spiders was not associated with any
further significant increase in osmolality (615 mosmol kg21) or
[Na+] (287 mmol l21), although these values were significantly
higher than those of hydrated fed spiders (515 mosmol kg21,
226 mmol l21 respectively). Potassium concentrations in the
haemolymph of dehydrated spiders were significantly elevated
by 1–2 mmol l21 in the dehydrated spiders compared with
hydrated spiders, although there was no significant effect of
feeding (Table 2).
It was not possible to obtain samples of stercoral fluid from
the dehydrated starved spiders. After 6 days of dehydration,
they had ceased to excrete urine and there was no fluid in the
stercoral pocket. Dehydration of feeding spiders raised the
osmolality of the stercoral fluid (620 mosmol kg21) above that
of both groups of hydrated spiders, but this value was not
significantly different from that of the haemolymph of the same
Rates of urine production and anal Na+ and K+ excretion
At the start of this experiment, following the pretreatment
of 3 weeks of starvation with water, both groups of spiders
produced urine at a mean rate of about 5 ml day21 (Fig. 1A),
as previously established for unfed hydrated spiders (Butt and
Taylor, 1986). Urine production decreased rapidly in the unfed
dehydrated spiders and, after 5 days, it had ceased. Fed spiders
produced urine throughout the 6 day dehydration period,
although the large post-prandial diuresis of more than
30 ml day21 previously observed in hydrated spiders (Butt and
Taylor, 1986) did not occur and the rate had declined to
1.8 ml day21 on the sixth day when, as noted above, few of the
spiders actually fed.
Spiders deprived of food and water excreted small quantities
of Na+ and K+ in the anal urine before its production ceased. In
the fed spiders, anal excretion of Na+ was similar, and also
declined, but K+ excretion was elevated throughout the 6 day
period (Fig. 1B,C). Individual measurements of the molar ratio
of Na+/K+ in the anal excreta of these spiders were highly
variable in both groups, but tended to increase in the unfed
group from about 0.5 to 1.3, as urine production slowed, and to
decrease to about 0.1 in the fed spiders (Fig. 1D). The mean
total quantities of ions excreted anally during the 6 day period
were 0.37 mmol Na+ and 0.57 mmol K+ (ratio 0.65) for starved
spiders, and 0.36 mmol Na+and 2.5 mmol K+ (ratio 0.14) in fed
spiders. The molar Na+/K+ ratio of the prey provided to the
spiders in these experiments was 0.47. Despite these differences
in the molar Na+/K+ ratio of the excreta and the food, the spiders
Table 2. The composition of the haemolymph and stercoral fluid of spiders starved or fed for 6 days with and without water
[Na+]
(mmol l–1)
Osmolality
(mosmol kg–1)
Condition
of spiders
Starved
Hydrated*
(N=8)
Dehydrated
(N=8)
Fed
Hydrated*
(N=9)
Dehydrated
(N=10)
[Cl−]
(mmol l–1)
[K+]
(mmol l–1)
Haemolymph Stercoral fluid Haemolymph Stercoral fluid Haemolymph Stercoral fluid Haemolymph Stercoral fluid
427±5
196±4
51±11c
5.1±0.2
64±12
156±5
579±15a
−
297±8b
−
7.0±0.1d
−
−
−
515±10
504±10
226±5
35±8c
4.5±0.3
125±6
225±3
78±8e
615±22a
620±6
287±19b
22.8±6c
6.4±0.2d
191±6
−
164±20
436±5
57±13e
*Hydrated values are from Butt and Taylor (1986); dehydrated values are series I spiders for which further data are given in Table 1.
Mean values with the same letter are not significantly different from each other. All other means in the same column are significantly different
(P<0.05; one-way ANOVA, Tukey–Kramer multiple comparisons test).
1142 A. G. BUTT
AND
H. H. TAYLOR
spiders (615 mosmol kg21). Despite being iso-osmotic, the
stercoral fluid of dehydrated spiders was markedly hypo-ionic,
the combined concentrations of the major inorganic ions, Na+,
K+ and Cl2, accounting for only about 60 % of its osmolality.
As previously noted for hydrated feeding spiders (Butt and
Taylor, 1986; Table 2), the Na+ concentration of the stercoral
fluid of dehydrated feeding spiders (22.8 mmol l21) was very
low and K+ was the predominant cation (191 mmol l21).
The effect of prior dehydration on post-prandial urine
production (series II)
In order to test whether the low urine output seen in
dehydrated spiders provided with prey (Fig. 1A) was primarily
associated with their dehydration or to the fact that their rate
of feeding progressively declined, spiders were dehydrated for
varying periods before being fed a single meal. The total
volume of urine produced over 3 days during and following
the meal was measured only in those spiders that were
observed to feed (Table 3). Hydrated spiders produced a total
of 54 ml of urine in the 3 day period. Removal of water and
simultaneous introduction of food reduced the diuresis to
19 ml. Dehydration prior to feeding resulted in a progressive
decline in the volume of urine until, after 96 h of dehydration
(about 10 % mass loss, 13 % water loss), the spiders only
produced 4 ml of urine following a meal comparable to that
consumed by the hydrated spiders. Thus, the magnitude of the
post-prandial anal diuresis appears to be influenced by the
dehydration state of the spider.
Ion and water balance of spiders dehydrated, fed a single
meal and then rehydrated (series III)
In hydrated spiders, a discrepancy between the composition
of the anal excreta and the ingested food was accounted for by
the excretion of a Na+-rich coxal fluid into the prey during
feeding (Butt and Taylor, 1986, 1991). To assess whether
dehydrated spiders similarly partition the excretion of ingested
ions, the anal and coxal excretion of Na+ and K+ were
measured for a single meal provided on day 4 of a 7 day period
of dehydration, which was subsequently followed by 3 days of
rehydration when drinking water was resupplied. A control
group was treated similarly but not fed.
During the first 3 days of dehydration, both groups
experienced about a 10 % decrease in body mass (13 % loss of
body water) (Fig. 2A). The starved spiders continued to lose
mass and had lost 18 % of their body mass (24 % water loss)
at the end of the 7 day period of dehydration. These spiders
regained all of this mass during the first day that water was
provided and maintained this mass for a further 2 days. The
other group was provided with prey for 1 day after 3 days of
dehydration. Body mass increased to slightly above the initial
value and, thereafter, continued to decline at a rate similar to
that observed before the meal. Provision of water for
rehydration resulted in a rapid increase in body mass to 114 %
of the original mass on the first day, with a subsequent decline
in mass over the next 2 days.
As seen above, dehydration resulted in a progressive decline
Table 3. The effect of water deprivation and varying degrees
of dehydration on the total volume of urine excreted by
spiders during 3 days following provision of a single meal
Duration of
dehydration
before feeding
% Loss of
initial mass
before feeding
Total volume
of urine
(ml)
0 h, hydrated,
water supplied
throughout (N=8)
0
54±6
0 h, hydrated but fed
without water (N=6)
0
19±8**
2.5±0.5
6.4±1.3
8.8±1.8
9.8±2.3
13±3***
12±4***
12±3**
4±1***
12 h (N=6)
48 h (N=8)
72 h (N=4)
96 h (N=10)
Values are means ± 1 S.E.M.
Series II spiders: the mean hydrated mass of the six groups ranged
from 0.71 to 0.78 g.
Significance levels of differences between the mean urine
production of the water-deprived groups and the group supplied with
water are: **P<0.01; ***P<0.001 (one-way ANOVA, Tukey–Kramer
multiple comparisons test).
For the five groups of water-deprived spiders, the volume of urine
produced (ml) was negatively correlated with their prior water loss (%
initial mass) (y=18.1−1.2x; P=0.013 that slope is zero; r=0.42) and
with the duration of the prior dehydration (days) (y=17.6−3.1x;
P=0.011 that slope is zero; r=0.43).
in anal urine production by starved spiders and a small diuresis
was associated with feeding during dehydration (Fig. 2B).
Both groups of spiders experienced an anal diuresis during the
first 2 days of rehydration, although the spiders fed during
dehydration produced more urine (2 day total, 66 ml) than the
starved controls (28 ml).
The brief anal diuresis during and following feeding was
responsible for a small increase in the anal excretion of Na+
(Fig. 2C) and a greater increase in the anal excretion of K+
(Fig. 2D). The net result was that, prior to feeding, the molar
Na+/K+ ratio of the excreta of both groups of spiders ranged
between 0.4 and 1.0, but in the 24 h following feeding the
molar Na+/K+ ratio of the excreta produced by the fed spiders
dropped to 0.15. Upon rehydration, there was an immediate
increase in anal K+ excretion, which was much greater in the
fed spiders and remained elevated for 2 days. Similar quantities
of Na+ were excreted by both groups on the first day of
rehydration, although anal Na+ excretion peaked in the fed
group on day 2.
Dehydrated spiders produced coxal fluid only while they
were actually feeding, and the radioactivity indicating coxal
excretion was located only in the prey debris (as previously
noted for continuously hydrated spiders; Butt and Taylor,
1991). Coxal fluid was responsible for the excretion of
2.6 mmol of Na+ and 0.112 mmol of K+ from their body pool
into the prey debris (Fig. 3). In contrast to continuously
hydrated spiders, rehydration of both fed and unfed spiders
Spider excretion 1143
Body mass (% of initial)
120
Dehydration
A
Fig. 2. Changes in body mass and in the anal excretion of water and
cations following a single meal in water-deprived Porrhothele
antipodiana. Fed spiders (open symbols) were provided with three
live cockroach nymphs for 1 day only (indicated by horizontal bar)
and then dehydrated (stippled bar) for a further 2 days before being
supplied with free water (hatched bar). The control group (filled
symbols) was starved throughout. (A) Percentage of initial total mass;
(B) daily production of anal urine; (C) daily anal excretion of Na+;
(D) daily anal excretion of K+. Points are means ± S.E.M.; N=14 for
dehydration period, N=6 for rehydration period for both groups. Series
III spiders; mean initial mass of both groups was 0.88 g.
Rehydration
Food
110
100
90
Anal urine production (ml day−1)
3
Anal K+ excretion (mmol day−1)
50
Anal Na+ excretion (mmol day−1)
80
Dehydration
B
40
Rehydration
Food
30
20
10
0
Dehydration
C
Rehydration
Food
2
1
0
4
Dehydration
D
Rehydration
Food
3
2
1
0
0
2
4
6
Time (days)
8
10
stimulated further coxal excretion onto the walls of the
container during the first 2 days. Control spiders excreted a
total of 0.67 mmol of Na+ and 0.248 mmol of K+ in this way
and spiders which had previously fed excreted a further
1.8 mmol of Na+ and 0.495 mmol K+ during the same period.
Summary of ion and water balance
The uptake of ions and water from the prey and their
excretion in the coxal and anal fluids in spiders fed for 1 day
during 7 days of dehydration, and subsequently supplied with
water (Figs 2, 3), is summarised in Table 4. 46 % (3.54 mmol)
of the ingested Na+ (7.63 mmol) was eliminated during the
dehydration period, mainly into the prey by the coxal organs,
during feeding. Only 19 % (3.16 mmol) of the ingested K+
(16.41 mmol) was excreted during this period, primarily via the
anal system. When the spiders were allowed to drink, a further
excretion equivalent to 44 % (3.38 mmol) of the ingested Na+
and 46 % (7.49 mmol) of K+ took place. Again, most of the Na+
was excreted in the coxal fluid and most of the K+ was excreted
anally.
Thus, at the end of the 10 day experimental period, the fed
spiders still retained about 9 % (0.71 mmol) of the Na+ from
the single meal and 35 % (5.76 mmol) of the K+. In contrast,
the unfed spiders experienced a small net loss of both ions
during the dehydration period (only anally) and a rather greater
loss upon subsequent rehydration by drinking (both anal and
coxal excretion). Although the overall molar ratio of Na+/K+
excretion was similar in both groups (about 0.6), this reflected
quite different handling of the two ions by the two excretory
sytems. In the unfed spiders, the anal system was the principal
route for excretion of both cations but, in fed spiders, the coxal
organs excreted most of the ingested Na+ and the anal system
most of the ingested K+.
It is estimated that 142 ml of water was obtained from the
prey. Of this, 45 ml was excreted during the dehydration period
in the anal (25 ml) and coxal (20 ml) fluids and 97 ml of fluid
was retained. On provision of drinking water, the secondary
anal and coxal diureses eliminated a further 98 ml of water.
Thus, total fluid excretion during the 10 day experimental
period (142.8 ml) approximated that obtained from the prey.
However, the correspondence is perhaps a little fortuitous. This
budget does not consider the much larger turnover associated
with evaporative water losses from the cuticle, the lungs and
the mouthparts during feeding, minimally estimated as
19 ml day21 (Butt and Taylor, 1986, 1991). This presumably
was replaced by drinking when water was resupplied.
Coxal excretion during dehydration (series IV)
The coxal organs therefore function in sodium excretion
during feeding in both hydrated and dehydrated spiders. In
hydrated spiders, the coxal fluid is markedly hypo-osmotic to
the haemolymph (Butt and Taylor, 1991). This might be
1144 A. G. BUTT
Coxal excretion of Na+
(mmol day−1)
4
AND
H. H. TAYLOR
Dehydration
Food
A
Table 4. A summary of the partitioning of the excretion of
water and ions between the coxal organs and the anal
system, during 7 days of dehydration (deprived of drinking
water) and 3 days of rehydration (water provided) in spiders
either starved continuously or provided with a meal on day 4
Rehydration
3
2
Starved
Water
(ml)
1
Total ingested
Coxal excretion of K+
(mmol day−1)
0.5
Dehydration
Food
B
0.4
Rehydration
0.3
0.2
0.1
0
0
2
4
6
Time (days)
8
10
Fig. 3. The effect of a single meal during dehydration on coxal
excretion of Na+ and K+ by Porrhothele antipodiana. Ion excretion
during the meal was calculated from differences in the dry mass and
ion content of the prey and debris and at other times by direct
measurement of loss to the paper lining of the chamber. (A) Daily
Na+ excretion; (B) daily K+ excretion. Series III spiders; other details
as for Fig. 2.
disadvantageous when no free water was available.
Consequently, we measured the rates of production and the
composition of the coxal fluid in spiders dehydrated to varying
levels.
Effect of dehydration on the rate of coxal excretion
As with hydrated spiders (Butt and Taylor, 1991), coxal
excretion by dehydrated spiders coincided with the start of
feeding and ceased with the completion of the meal. Increasing
dehydration was associated with a progressive decline in the
rate of coxal fluid production, by single anterior coxal
openings, from about 19 ml h21 g21 body mass in fully
hydrated spiders to about 4 ml h21 g21 body mass after 4 days
of dehydration and 11 % loss of body mass (14.7 % water loss)
(Fig. 4A). Spiders were never observed to feed without
producing coxal fluid. It was not possible to induce restrained
spiders to feed if they had lost more than 12 % of their body
mass (16 % water loss) through dehydration.
The effect of dehydration on the composition of the coxal
fluid and haemolymph
The osmolality, [Na+] and [K+] of the haemolymph
increased with increasing dehydration (Fig. 4B–D). The
osmolality and [Na+] of the coxal fluid were lower than those
of the haemolymph until the spiders had lost more than 4 % of
their body mass through dehydration. Thereafter, the coxal
0
Na+
Fed
K+
(mmol) (mmol)
142.0
7.63
16.41
20.0
25.0
45.0
2.63
0.91
3.54
0.11
3.05
3.16
Excretion during rehydration period
Coxal
5.70
0.67
0.25
Anal
39.1
0.69
1.88
Total
44.8
1.36
2.13
18.2
79.6
97.8
1.80
1.58
3.38
0.50
6.99
7.49
38.2
104.6
142.8
4.43
2.49
6.92
0.61
10.04
10.65
0.67
1.16
1.83
0
Na+
K+
(mmol) (mmol)
Excretion during dehydration period
Coxal
0
0
0
Anal
10.1
0.48
1.01
Total
10.1
0.48
1.01
Total excretion
Coxal
5.70
Anal
49.2
Total
54.8
0
Water
(ml)
0.248
2.89
3.14
Calculated from data presented in Figs 2 and 3.
Series III spiders: the mean initial mass of both groups was 0.88 g
(N=14 for the dehydration period and N=6 for the rehydration period
in both cases).
fluid was either iso-osmotic or slightly hyperosmotic to the
haemolymph. Accompanying the increase in osmolality was a
parallel increase in the [Na+] of the fluid, which became
essentially iso-ionic with the haemolymph (Fig. 4C). In
contrast, there was no obvious trend in the [K+] of the coxal
fluid (Fig. 4D), which was 4–6 times the [K+] of the
haemolymph at all levels of dehydration.
Discussion
As spiders are, in effect, fluid feeders, it might be expected
that an additional water source would be unnecessary when
feeding ad libitum. However, for this to be true they would
need to produce excretory fluids more concentrated than the
food source with respect to all ions, in order to offset
evaporative losses that occur during feeding and the period
before the next meal. The present study indicates that this is
not the case in P. antipodiana, although both the coxal and the
anal excretory systems do exhibit regulatory, water-conserving
responses to dehydration involving the rate of production and
the composition of the anal urine and the coxal fluid.
Hydrated and feeding P. antipodiana of about 1 g body mass
deliver coxal fluid into the prey at 19 ml h21 from each coxal
opening (approximately 75 ml h21 from all four coxal
openings). The rate of coxal excretion during feeding was
reduced by nearly 80 % (4 ml h21 from each coxal opening)
after 4 days of dehydration and the loss of 11 % of the body
Rate of coxal excretion
(ml h−1 organ−1 g−1 body mass)
Spider excretion 1145
A
25
20
15
10
5
0
0
2.5
5.0
7.5
10.0
12.5
Osmolality
(mosmol kg−1 H2O)
800 B
600
400
Haemolymph
Coxal fluid
200
0
0
2.5
5.0
7.5
10.0
12.5
[Na+] (mmol l−1)
400 C
300
200
Haemolymph
Coxal fluid
100
0
0
5.0
7.5
10.0
12.5
D
80
[K+] (mmol l−1)
2.5
Haemolymph
Coxal fluid
60
40
20
0
0
2.5
5.0
7.5
10.0
Loss of body mass (% of initial)
12.5
Fig. 4. Effect of progressive dehydration (as percentage loss of body
mass) on coxal fluid production by restrained Porrhothele
antipodiana during feeding and on the composition of the coxal fluid
and the haemolymph. (A) The rate of coxal fluid production, by single
anterior coxal organs, normalised to the initial total mass of the spider;
(B) the osmolality of coxal fluid and haemolymph; (C) [Na+] of coxal
fluid and haemolymph; (D), [K+] of coxal fluid and haemolymph.
Spiders were dehydrated at room temperature and humidity.
Haemolymph (filled symbols) and coxal fluid (open symbols)
obtained from same spider are linked by vertical lines. Values are
from individual spiders. Trend lines are fitted by least-squares linear
regression to all values in A and to haemolymph values only in B–D.
mass (Fig. 4). This reduction in coxal excretion might simply
reflect impaired ultrafiltration consequent on a lowered internal
volume and pressure, although other mechanisms are clearly
possible. Similarly, the production of anal urine was reduced.
Hydrated spiders excreted 54 ml of urine over 3 days in a
diuresis following a single meal. In contrast, the post-prandial
diuresis in spiders dehydrated for 4 days was reduced to only
4 ml (Table 3). Starved spiders ceased to excrete urine after 5
days of dehydration and fluid was absent from the stercoral
pocket. As spiders desiccated to this degree did not feed, fluid
excretion was completely inhibited at this point.
The major reduction in the anal diuresis was associated with
the lack of drinking water and occurred before the spiders had
time to become significantly dehydrated. Provision of prey to
hydrated spiders and simultaneous removal of water reduced
the post-prandial diuresis by two-thirds to 19 ml. However, a
delayed post-prandial diuresis of normal size occurred, even in
markedly dehydrated spiders, on resupplying drinking water
(Fig. 2). This implies that the anal diuresis must be supported
by drinking after the meal and is consistent with a previous
inference (Butt and Taylor, 1986) that the main function of the
diuresis is to eliminate ions ingested with the meal rather than
to regulate the internal volume change. Thus, reductions in the
volumes of the urine and the coxal fluid allowed dehydrated
spiders to gain about 100 ml of water from a single meal
(Table 4), potentially offsetting the transpirational losses of
around about 20 ml day21 (Butt and Taylor, 1986). However,
this fluid was not without cost as it was associated with a
significant ion load requiring later excretion.
Consideration of the ion balance of dehydrated spiders that
were given a single meal (Table 4) indicates that, as previously
observed in hydrated spiders (Butt and Taylor, 1991), the coxal
organs are the dominant route for the elimination of Na+, being
responsible for 74 % of excretion of this ion in the dehydration
period (2.63 mmol of a total 3.54 mmol). In contrast, the anal
system excreted most of the K+ (97 % or 3.05 mmol of
3.16 mmol). Importantly, both the anal system and the coxal
organs were rather ineffective in eliminating the ingested ions
under dehydrating conditions. Only 46 % of 7.63 mmol of
ingested Na+ and 19 % of the 16.41 mmol of ingested K+ were
excreted in the 3 days after the meal. On rehydration, a similar
quantity of Na+ and a rather larger quantity of K+ were
excreted, with a similar partitioning between the coxal and anal
fluids being maintained.
The effects of dehydration and rehydration on the excretion
of the water and salts obtained from a single meal are
summarized in Fig. 5 and compared schematically with a
comparable experiment on hydrated spiders (Butt and Taylor,
1991). Over the 10 day period, the total inputs and outputs,
here expressed as percentages of the quantities ingested from
the prey, are similar in the two situations, as is the partitioning
of Na+, K+ and water excretion between coxal and anal routes.
As noted already, the main effect of dehydration is that the
excretion of all three components, by both routes, is largely
delayed until secondary diureses, which occur only when
drinking water again becomes available. The adaptive
significance of this presumably relates to the over-riding need
to maintain body volume and turgor for locomotion.
Significantly, the second coxal diuresis is released to the
substratum, suggesting a sole excretory role, whereas, in
1146 A. G. BUTT
AND
H. H. TAYLOR
A
EVAPORATION
Water 128 %
HYDRATED
(water available
ad libitum)
Midgut diverticula
Malpighian tubules
Sucking stomach
DRINKING
Water 120 %
WATER
Stercoral pocket
INGESTED
8.3 mmol
17.2 mmol
Water 148 ml
Na+
K+
Coxal organs
ANAL URINE
COXAL FLUID
Release to prey
Na+
63 %
K+
8%
Water 28 %
After feeding
Na+
K+
Water
EVAPORATION
Water 136 %
B
DEHYDRATED
(water available
on days 8-10)
and drinking
31 %
46 %
64 %
Midgut diverticula
Malpighian tubules
Sucking stomach
Stercoral pocket
DRINKING
WATER
Water
137 %
INGESTED
7.6 mmol
16.4 mmol
Water 142 ml
Na+
K+
Coxal organs
ANAL URINE
COXAL FLUID
Release to prey + Release to substratum = Total
Na+
34 %
24 %
58 %
K+
1%
3%
4%
Water 14 %
13 %
37 %
After feeding + After drinking = Total
Na+
12 %
21 %
33 %
K+
19 %
43 %
61 %
Water 18 %
56 %
74 %
Fig. 5. Summary and comparison of the partitioning of ingested cations and water between the two excretory fluids, coxal fluid and anal urine, in
(A) hydrated and (B) dehydrated spiders, Porrhothele antipodiana. In A, previously starved, hydrated spiders, weighing about 1 g, were provided
with drinking water ad libitum for 10 days and a single meal of three cockroach nymphs (about 0.45 g total) on day 2 (data from Table 3 in Butt
and Taylor, 1991). Budget B encompasses a 10 day period in which spiders were deprived of water for 7 days, provided with a similar meal on
day 4, and subsequently allowed to drink on days 8–10 (present study, series III, Table 4). Excretion of ions and water during the dehydration/feeding
period, and during rehydration, are totalled separately. In both, the inputs and outputs of ions and water are expressed as percentages of the quantities
ingested from the meal (quite similar, as shown). Evaporative water losses and drinking (obtained by difference) are minimal values that assume
overall water balance and similar transpiration rates to those of unfed spiders (Butt and Taylor, 1986). In both groups, Na+ is eliminated principally
in the coxal fluid whereas K+ is excreted mainly in the anal urine. When water is available (A), coxal fluid is discharged only into the prey during
feeding (probably also serving a mechanical role in ingestion), and excreted ions are left behind in the prey debris. Anal urine is produced in a
post-prandial diuresis, peaking 1–2 days after the meal. In dehydrated spiders deprived of water (B), the overall budgets are similar but the excretion
of a large fraction of the cations is delayed until drinking water becomes available. In the secondary diureses, both coxal and anal fluids are deposited
on the substratum. The relatively high evaporative losses indicate that these spiders must drink to maintain water balance and that the primary
function of the diureses relates to iono-regulation and not volume regulation. Deficits in the excreted cations represent net gain by the spiders.
Spider excretion 1147
hydrated spiders, all is delivered into the prey and may also aid
ingestion as discussed previously (Butt and Taylor, 1991).
Minimal volumes of water imbibed in the 10 day period were
calculated assuming that the spiders are in overall water
balance (in practice, there is a small net gain of water
associated with feeding) and that evaporation is similar to that
in non-feeding spiders (Butt and Taylor, 1986; in practice,
feeding activity might increase respiratory losses and there
would be losses from the prey). Interestingly, the volumes of
all four major components of water balance in the 10 day
period, i.e. ingestion, drinking, evaporation and excretion, in
both groups are all of similar magnitude. All are obligatory and
it appears that the spider has little flexibility to reduce any of
them.
The excretion of ions and water by routes other than the
Malpighian tubules and anus occurs in other arachnids. In the
argasid tick Ornithodorus moubata, the coxal glands function
in volume and osmotic regulation, excreting ions to the exterior,
during and after feeding (Lees, 1946; Kaufman et al. 1981,
1982). In the blood-sucking gamasid mite Ornothonysus bacoti,
and the ixodid ticks Boophilus microplus and Dermacentor
andersoni, the salivary glands excrete electrolytes and fluid into
the host (Belozerov, 1958; Tatchell, 1967, 1969; Kaufman and
Phillips, 1973a,b) in analogous fashion to coxal excretion
during feeding in hydrated P. antipodiana. In fact, the summary
budget presented for D. andersoni (Kaufman and Phillips,
1973a) demonstrates a partitioning of excreted cations rather
similar to that of P. antipodiana (Table 4; Fig. 5); most of the
Na+ is excreted by the salivary glands into the host during
feeding, while K+ is eliminated via the anus. However, it also
important to emphasise the differences between the spider and
the tick. (1) The coxal glands are thought to generate fluid by
ultrafiltration (Butt and Taylor, 1991), whereas salivation in the
tick occurs by secretion (Kaufman and Phillips, 1973b). (2) The
large size of the vertebrate host ensures that the tick’s salivary
excretion is diluted and washed away, whereas most of the
spider’s coxal fluid must be recycled via the mouth (Table 4
and Fig. 5 record only the net quantities left behind in the prey).
(3) Evaporative losses are negligible in relation to the water
turnover in the feeding tick, but they are substantial in P.
antipodiana. Thus, salivation in the tick is important in
elimination of both ions and water but, in the spider, water
excretion by both routes should perhaps be viewed as
unavoidable consequences of ion and nitrogen excretion, and of
the feeding process itself. (4) The tick does not need to drink
fresh water but the spider does.
The selective advantage of K+-handling by the anal system
and Na+-handling by the coxal system is unclear. Possibly the
coxal organs, although demonstrating some iono-regulatory
function, are primarily retained in relation to ingestion of prey.
Given the high [Na+] of the haemolymph, and the high rates
of coxal fluid production by ultrafiltration, Na+ elimination is
clearly favoured. By default, the function of K+ elimination
and final adjustment of Na+/K+ balance would then fall on the
anal system. The origin of the anal urine in the Malpighian
tubules or midgut diverticula of spiders has not yet been
demonstrated, but the comparison with the K+-coupled fluid
secretion of most insect Malpighian tubules (Phillips, 1981) is
tempting. Maddrell (1981) speculates that slowly operating,
secretion-based Malpighian tubule systems may have evolved
in insects in relation to their small size and xeric habitats.
Similar arguments could apply to spiders. It should be noted
that the K+-rich anal excretion of the tick D. andersoni was
considered more likely to result from the lack of absorption of
K+ in the gut rather than from their active excretion by the
Malpighian tubules (Kaufmann and Phillips, 1973a). Such a
mechanism cannot be excluded in P. antipodiana, although
ready assimilation of 42K by the the spider Lycosa punctata
(Van Hook, 1971) argues against this mechanism.
The reduced excretion of ingested ions under dehydrating
conditions reflects the inability of the coxal organs and the anal
system to produce fluids that are appreciably hyperosmotic to
the haemolymph. In hydrated spiders, the coxal fluid is hypoosmotic to the haemolymph and the stercoral fluid, although
iso-osmotic, is significantly hypo-ionic. The concentrations of
Na+, K+ and Cl2 fully account for the osmolality of the
haemolymph, but represent only 47 % of the stercoral fluid
osmolality. The additional osmolytes have not yet been
identified. In response to dehydration, the osmolality of the
coxal fluid increased from about 200 mosmol kg21 to isoosmotic or slightly hyperosmotic to the haemolymph (about
450 mosmol kg21) when the spiders had lost 6 % of their live
mass. The osmolality of the stercoral fluid increased slightly
with dehydration, but remained approximately iso-osmotic,
and still markedly hypo-ionic, to the haemolymph.
The inability of either the coxal organs or the anal system
to excrete fluid hyperosmotic to the haemolymph prevents P.
antipodiana from maintaining positive fluid balance when fed
ad libitum without drinking water. Unfed dehydrated spiders
experienced increases in the concentrations of total body ions
and the osmolality and concentrations of haemolymph ions
when compared with hydrated starved spiders. Feeding during
dehydration resulted in even greater increases in ion
concentrations as a result of the retention of K+ and Na+
ingested with the prey (Table 1). This retention of prey ions
may be a factor leading to cessation of feeding during
dehydration. Another factor may be that reduced coxal fluid
output (Fig. 4A) eventually interferes with the process of
maceration and ingestion of the prey.
The inability of P. antipodiana to produce a hyperosmotic
fluid may limit its distribution to terrestrial habitats with at
least intermittent access to fresh or brackish water. In this
respect, P. antipodiana resembles terrestrial crustaceans,
which also do not produce concentrated urine (Edney, 1968;
Bliss, 1968; Harris and Kormanik, 1981; Taylor et al. 1993).
P. antipodiana is always found in hygric environments and
often constructs a silk-lined tunnel extending up to 40 cm down
among shingle and stones (Todd, 1945). Rain and dew collect
on the web and probably provide the main source of drinking
water. When weather conditions prevent precipitation for
several days, it appears that foraging and feeding would be
impaired.
1148 A. G. BUTT
AND
H. H. TAYLOR
When spiders were provided with drinking water after a
period of dehydration they rapidly regained mass. Both fed and
unfed spiders experienced a period of diuresis involving both
the anal system and the coxal organs. Hydrated starved spiders
produce quite constant quantities of urine each day in which
small quantities of body salts, and presumably nitrogenous
products associated with metabolism, were eliminated (Butt
and Taylor, 1986). Dehydrated starved spiders progressively
reduced the volume of urine excreted daily and, after a period,
they ceased to excrete urine at all. It is therefore likely that the
diuresis seen in the starved spiders during rehydration was
involved in elimination of metabolic products accumulated
during the period of dehydration. Indeed, the quantity of ions
excreted by the rehydrating spiders following 7 days of
dehydration (1.36 mmol Na+ and 2.13 mmol K+) was similar to
the total excretion of Na+ and K+ by hydrated starved spiders
over the same period (average daily excretion of hydrated
starved spiders 0.2 mmol Na+ and 0.3 mmol K+; Butt and
Taylor, 1986).
The diuresis during rehydration was greater in fed spiders
than in starved spiders, and the spiders that had been fed
excreted greater quantities of Na+ and K+. This presumably
completes the elimination of excess ions ingested with the
meal. When faced with a meal during dehydration, P.
antipodiana either kills the prey and does not feed or consumes
the prey and permits changes in ion content and composition
of the body fluids and tissues. These changes are tolerated until
such time as drinking water is available to support anal and
coxal excretion of the excess salts.
The observation that coxal fluid was excreted during
rehydration is a further demonstration of coxal excretion at a
time other than during the actual feeding process. Previously
it has been shown that the somewhat artificial conditions of
very high dietary or injected salt loading also induced coxal
excretion to the substratum rather than returning it to the prey
(Butt and Taylor, 1991). The present example, under
conditions that P. antipodiana is more likely to encounter, is
consistent with the excretory role proposed for the coxal organs
in this spider. The variations in rates of coxal excretion and
changes in composition of coxal fluid with dehydration are also
consistent with a regulatory role. Dehydration to 11 % mass
loss was associated with an 80 % reduction in the rate of coxal
excretion induced by feeding. Accompanying the decrease in
rate of coxal excretion was an increase in both the osmolality
and the Na+ concentration of the coxal fluid. Despite a much
greater relative increase in haemolymph [K+] (78 %) than
[Na+] (25 %), there was no change in the K+ concentration of
the coxal fluid. This is consistent with the postulated role of
the coxal organs in Na+ regulation and with previous
observations that they did not respond to modified dietary K+
intake (Butt and Taylor, 1991).
P. antipodiana was chosen for these studies partly because
it was a representative of the more primitive Mygalomorpha
and it was expected to exhibit more generalised features than
the Araneomorpha or true spiders. It is now important to
determine whether other representatives of both groups exhibit
comparable excretory mechanisms. As yet, there have been no
critical studies on excretory regulation in any araneomorph.
Many of these spiders inhabit xeric environments seemingly
devoid of water, e.g. cobweb spiders (Theridiidae) in dry
buildings. While it is theoretically possible that such spiders
could trek long distances to water, or select hypo-osmotic prey,
or take up atmospheric water vapour, there is no direct
evidence to support the use of these strategies. Clearly, it is
still important to investigate whether a capability for
production of hyperosmotic excretory fluids is a feature of
spider osmoregulation.
References
BELOZEROV, U. N. (1958). Influence of humidity on Ornithonysus
bacoti mites (Parasitiformes, Liponyssidae). Ent. Rev. 37, 36–49.
BLISS, D. E. (1968). Transition from water to land in decapod
crustaceans. Am. Zool. 8, 355–392.
BURSELL, E. (1957). Spiracular control of water loss in the tsetse fly.
Proc. R. ent. Soc. Lond. 32, 21–29.
BUTT, A. G. (1983). Salt and water balance of the spider, Porrhothele
antipodiana (Mygalomorpha: Dipluridae). PhD thesis, University
of Canterbury, Christchurch, New Zealand. 185pp.
BUTT, A. G. AND TAYLOR, H. H. (1986). Salt and water balance of the
spider, Porrhothele antipodiana (Mygalomorpha: Dipluridae):
effects of feeding upon hydrated animals. J. exp. Biol. 125, 85–106.
BUTT, A. G. AND TAYLOR, H. H. (1991). The function of spider coxal
organs: effects of feeding and salt loading on Porrhothele
antipodiana (Mygalomorpha: Dipluridae). J. exp. Biol. 158,
439–461.
BUXTON, B. H. (1913). The coxal glands of arachnids. Zool. Fahrb.
Abt. Anat. 14, 231–282.
CLARKE, K. U. (1979). Visceral anatomy and arthropod phylogeny. In
Arthropod Phylogeny (ed. A. P. Gupta), pp. 467–549. New York:
Van Nostrand Rheinhold Co.
CLOUDSLEY-THOMPSON, J. L. (1957). Studies in diurnal rhythms. V.
Nocturnal ecology and water relations of the British cribellate
spiders of the genus Ciniflo Bl. Linn. Soc. J. Zool. 43, 134–152.
DAVIES, M. E. AND EDNEY, E. B. (1952). The evaporation of water
from spiders. J. exp. Biol. 29, 571–582.
EBELING, W. (1974). Permeability of insect cuticle. In The Physiology
of Insecta, vol. VI (ed. M. Rockstein), pp. 271–343. New York:
Academic Press.
EDNEY, E. B. (1968). The effect of water loss on the haemolymph of
Areniviga sp. and Periplaneta americana. Comp. Biochem. Physiol.
25, 149–158.
GILBY, A. R. (1980). Transpiration, temperature and lipids in insect
cuticle. Adv. Insect Physiol. 15, 1–34.
GOODRICH, F. S. (1945). The study of nephridial and genital ducts
since 1895. Q. Jl microsc. Sci. 86, 113–142.
HADLEY, N. F. (1978). Cuticular permeability and lipid composition
of the black widow spider, Latrodectus hesperus. Symp. zool. Soc.
Lond. 42, 429–438.
HADLEY, N. F. (1994a). Water Relations of Terrestrial Arthropods.
San Diego: Academic Press.
HADLEY, N. F. (1994b). Ventilatory patterns and respiratory
transpiration in adult terrestrial insects. Physiol. Zool. 67, 175–189.
HADLEY, N. F. AND QUINLAN, M. C. (1989). Cuticular permeability of
Spider excretion 1149
the black widow spider Latrodectus hesperus. J. comp. Physiol. B
159, 243–248.
HARRIS, R. R. AND KORMANIK, G. A. (1981). Salt and water balance
and antennal gland function in three Pacific species of terrestrial
crab (Gecarcoidea lalandii, Cardisoma carnifex, Birgus latro). II.
The effects of desiccation. J. exp. Zool. 218, 107–116.
HUMPHREYS, W. F. (1975). The influence of burrowing and
thermoregulatory behaviour on the water relations of Geolycosa
godeffroyi (Araneae: Lycosidae), an Australian wolf spider.
Oecologia 21, 291–311.
KAUFMAN, S. E., KAUFMAN, W. R. AND PHILLIPS, J. E. (1981). Fluid
balance in the argasid tick, Ornithodorus moubata, fed on modified
blood meals. J. exp. Biol. 93, 243–256.
KAUFMAN, S. E., KAUFMAN, W. R. AND PHILLIPS, J. E. (1982).
Mechanism and characteristics of coxal fluid excretion in the
argasid tick Ornithodorus moubata. J. exp. Biol. 98, 343–352.
KAUFMAN, S. E. AND PHILLIPS, J. E. (1973a). Ion and water balance
in the ixodid tick, Dermacentor andersoni. I. Routes of ion and
water excretion. J. exp. Biol. 58, 523–536.
KAUFMAN, S. E. AND PHILLIPS, J. E. (1973b). Ion and water balance
in the ixodid tick, Dermacentor andersoni. II. Mechanism and
control of salivary excretion. J. exp. Biol. 58, 537–547.
KRASFUR, E. S. (1971a). Behaviour of thoracic spiracles of Aedes
mosquitoes in controlled relative humidities. Ann. ent. Soc. Am. 64,
93–97.
KRASFUR, E. S. (1971b). Influence of age and water balance on
spiracular behaviour in Aedes mosquitoes. Ann. ent. Soc. Am. 64,
97–102.
LEES, A. D. (1946). Chloride regulation and the function of coxal
glands in ticks. Parasitology 37, 172–184.
LOVERIDGE, J. P. (1968). The control of water loss in Locusta
migratoriodes R. and F. II. Water loss through the spiracles. J. exp.
Biol. 49, 15–29.
MADDRELL, S. H. P. (1981). The functional design of the insect
excretory system J. exp. Biol. 90, 1–15.
MILLER, P. (1964a). Factors altering spiracular control in adult
dragonflies: water balance. J. exp. Biol. 41, 331–343.
MILLER, P. (1964b). Factors altering spiracular control in adult
dragonflies: hypoxia and temperature. J. exp. Biol. 41, 345–357.
MILLOT, J. (1949). Ordre des Araneides (Araneae). In Traité de
Zoologie, vol. VI (ed. P. P. Grassé), pp. 588–743. Paris: Masson et
Cie.
PAUL, R. AND FINCKE, T. (1989). Booklung function in arachnids. II.
Carbon dioxide release and its relations to respiratory surface, water
loss and heart frequency. J. comp. Physiol. B 159, 419–432.
PHILLIPS, J. E. (1964). Rectal absorption in the desert locust,
Schistocerca gregaria Forskal. I. Water. J. exp. Biol. 41,
15–38.
PHILLIPS, J. E. (1970). Apparent transport of water by insect excretory
systems. Am. Zool. 10, 413–416.
PHILLIPS, J. E. (1981). Comparative physiology of insect renal
function Am. J. Physiol. 241, R241–R257.
RAMSAY, J. A. (1952). The excretion of sodium and potassium by the
Malpighian tubules of Rhodnius. J. exp. Biol. 29, 110–126.
RAMSAY, J. A. (1955). The excretory system of the stick insect,
Dixippus morosus (Orthoptera, Phasmidae). J. exp. Biol. 32,
183–190.
RAMSAY, J. A. (1964). The rectal complex of the meal worm Tenebrio
moliter L. (Coleoptera, Tenebrionidae). Phil. Trans. R. Soc. B 248,
279–314.
ROBINSON, G. L. AND PAIM, U. (1969). Regulation of external
respiration by the booklung spiracles of the spiders Araneus
diadematus Clerck and A. marmoreus Clerck. Can. J. Zool. 47,
355–364.
SEYMOUR, R. S. AND VINEGAR, A. (1973). Thermal relations, water
loss and oxygen consumption of a North American tarantula.
Comp. Biochem. Physiol. 44, 83–96.
STEWART, D. M. AND MARTIN, A. W. (1970). Blood and fluid balance
of the common tarantula, Dugesiella hentzi. Z. vergl. Physiol. 70,
223–246.
TATCHELL, R. J. (1967). Salivary secretion of the Cattle Tick as a
means of water elimination. Nature 213, 940–941.
TATCHELL, R. J. (1969). The ionic regulatory role of the salivary
secretion of the cattle tick Boophilus microplus. J. Insect Physiol.
15, 1421–1430.
TAYLOR, H. H., GREENAWAY, P. AND MORRIS, S. (1993). Adaptations
to a terrestrial existence by the robber crab Birgus latro L. VIII.
Osmotic and ionic regulation on freshwater and saline drinking
regimens. J. exp. Biol. 179, 93–113.
TODD, V. (1945). Systematic and biological account of the New
Zealand Mygalamorpha (Arachnida). Trans. R. Soc. N.Z. 74,
375–407.
VAN HOOK, R. I. (1971). Energy and nutrient dynamics of spider and
orthopteran populations in a grassland ecosystem. Ecol. Monogr.
41, 1–26.
WIGGLESWORTH, V. B. (1931). The physiology of excretion in a blood
sucking insect, Rhodnius prolixus (Hemiptera, Reduviidae). I. The
composition of the urine. J. exp. Biol. 8, 411–427.
WIGGLESWORTH, V. B. (1945). Transpiration through the cuticle of
insects. J. exp. Biol. 21, 97–114.