Aspects of excretion of antlion larvae (Neuroptera:

(1)

Aspects of excretion of antlion larvae (Neuroptera:

myrmeleontidae) during feeding and non-feeding periods

Amanda Van Zyl

^*1

, M.C. Van Der Westhuizen, T.C. De K. Van Der Linde

Department of Zoology-Entomology, University of the Free State, P.O. Box 339, Bloemfontein, 9300, South Africa Received 20 March 1998; received in revised form 15 June 1998

Abstract

The main nitrogenous excretory products were determined for third instar Cueta sp. and Furgella intermedia larvae during periods of food abundance and for F. intermedia during starvation periods. Biochemical analysis indicated that allantoin was the main nitrogenous excretory product, with smaller quantities of ammonia, urea and uric acid. Respectively 9 and 13 amino acids of low concentrations (0.005–0.329 g/l) were detected by high pressure liquid chromatography in the excreta of Cueta sp. and F. intermedia larvae. The volume of urine produced and concentrations of the nitrogenous excretory products of fed Cueta sp. and fed F. intermedia larvae did not differ. F. intermedia excreted smaller volumes of urine and smaller quantities of nitrogenous excretory products during starvation than during periods of food abundance. Feeding conditions rather than the pitbuilding or non-pitbuilding lifestyles seem to be the major influence on the excretory products of these antlion larvae.1998 Elsevier Science Ltd. All rights reserved.

Keywords: Neuroptera; Myrmeleontidae; Excretion; Allantoin; Water

1. Introduction

A strong correlation usually exists between the major nitrogenous excretory products and the nature of an insect’s environment, where aquatic forms often excrete ammonia and terrestrial forms uric acid (Cochran, 1985).

The antlion larvae, Cueta sp. and Furgella intermedia (Markl), live in the semi-arid to arid Kalahari desert (28°21⬙E, 21°16⬙S), where food resources are unpredict- able for these larvae (Van Zyl et al., 1996). As surface water is often absent in the natural environment of these antlion larvae, it has been suggested that food is their primary water resource during the dry season (Van Zyl, 1995). Antlion larvae are also unable to take up atmos- pheric water vapour as demonstrated for Myrmeleon medialis Banks, Cueta lanceolatus Navas and Syngenes longicornis (Rambur)(Youthed, 1973). The pitbuilder Cueta sp. and non-pitbuilder F. intermedia larvae should

* Corresponding author. Fax: 00 27 12 3625242; e-mail: avanzyl@- scientia.up.ac.za

1Present address: 244 Carinus Street, Meyerspark, 0184, South Africa. Fax: (27 12) 3625242; E-mail: avanzyl@scientia.up.ac.za.

PII: S 0 0 2 2 - 1 9 1 0 ( 9 8 ) 0 0 1 0 0 - 0

therefore be able to conserve water e.g. by excreting uric acid, during prolonged periods of food shortage.

On the other hand, antlion larvae are extra-intestinal digesters, i.e. they inject enzymes and probably ‘poison’

(Gaumont, 1976) into their prey, which dissolves the soft internal tissue of the prey. The fluid is then drawn out of the prey and into the alimentary canal of the antlion larva. Van Zyl et al. (1997) demonstrated that third instar Cueta sp. larvae (body weight of 31 mg) extracted 73%

of the total wet weight of sixth instar Hodotermes mos- sambicus (Hagen) larvae (body weight of 20 mg).

Antlion larvae ingest therefore large quantities of fluid during extra-intestinal digestion and a need exists for the storage or elimination of large quantities of excess fluid after feeding. This was demonstrated in other fluid feed- ers, e.g. bloodsucking and plant-sucking insects where an abundance of water is present after a meal and the urine voided is a crystal clear fluid (Wigglesworth, 1965).

The study on the excretory products of antlion larvae is to a large extent facilitated in that the alimentary canal is discontinuous (Lozı´nski, 1911; Van Zyl, 1995). No contamination by faeces occurs in the urine and the excretory products are primarily of metabolic origin.

Unfortunately, only a limited number of studies have

(2)

been conducted on the excretion of neuropteran species.

Shaw (1955) and Staddon (1955) conducted an in depth study on the ionic regulation of the larvae of the aquatic neuropteran, Sialis lutaria L. Ammonia is the main nitrogenous excretory product in these larvae. Razet (1961, quoted by Bursell, 1967) demonstrated that uric acid dominated in the excreta of the antlion larva, Uro- leon nostras (Fourcroy), with smaller quantities of allan- toin present. Spiegler (1962b) observed urate storage in the larvae of the green lacewing, Chrysopa carnea Steph. Of all these examples only U. nostras occurred in a terrestrial environment comparable to that of Cueta sp. and F. intermedia larvae.

In the present study the main nitrogenous excretory products (uric acid, allantoin, urea, ammonia and amino acids) of Cueta sp. and F. intermedia larvae were ident- ified. Excretion during periods of food abundance and food shortage was then related to survival in the semi- arid to arid Kalahari desert.

2. Materials and methods

Third instar Cueta sp. and F. intermedia larvae were collected in the Kalahari desert and transferred to glass vials (120 mm in depth⫻ 20 mm in diameter) filled to a depth of 60 mm with sterilised sand from their natural environment. Antlion larvae were acclimated to 28 ⫾ 2°C with a diel cycle of 12:12 (L:D). Termites, sixth instar H. mossambicus larvae, were used as their food supply.

2.1. Uric acid, allantoin, ammonia and urea determinations

Antlion larvae were starved for 42 days prior to the experiments. Excretory products were collected from fed and starved larvae in two treatments. In the first treat- ment nine Cueta sp. (body weight 29.8 ⫾ 7.0 mg) and 14 F. intermedia larvae (body weight 52.9⫾ 10.0 mg) were each fed one termite (approx. 12 mg) every second day over a 14 d period. These are referred to as ‘fed larvae’. The excreta of the individual antlion larvae and the termite carcasses were collected at the next feeding and frozen at ⫺ 12°C. Presence of excreta and wet weight of termites before feeding were recorded every second day. Excreta was thus collected on seven of the 14 experimental days and pooled for each individual.

The liquid excreta forms pellets in the sterile sand. These could be removed by sieving the sand through a 2 mm mesh screen. Prey carcasses were removed before sieving. As antlion larvae eject the prey carcass above the soil surface within an hour after feeding and excreted the urine several hours later in the sand, at a depth of 20 mm below the soil surface where the antlion larvae

were situated, it is unlikely that the prey carcass could have contaminated the excreta in any way.

In the second treatment in total 16 F. intermedia lar- vae were each fed one termite of approximately 19 mg once. These larvae are referred to as ‘starved larvae’.

The excreta was collected 10 times (on day 4, 5, 6, 8, 10, 12, 14, 18, 25 and 36) over a 36 d period and pooled for each individual. Presence of excreta was noted on the collecting days.

The uric acid, allantoin, ammonia and urea concentrations were determined for the excreta of the fed and starved antlion larvae, respectively. The volume of urine excreted by the antlion larvae was estimated by determining the quantity of water that saturated a known weight of dry sterilised Kalahari sand viz. 0.1018 ⫾ 0.017 ml of water/g dry sand (n ⫽ 9). The dry weight of the collected pellets (urine mixed with sterile sand) was then used to estimate the volume of urine excreted by the antlion larvae.

For analysis the collected dry excreta of fed larvae and of starved larvae were homogenised in 1.0 ml and 0.5 ml of 0.6% Li2CO3, (pH 11.53) respectively. All nitrogenous products tested, are soluble in this substance. The antlion excretory extract was heated for 10 min at 100°C, centrifuged for 10 min at 4 000 g and kept at 4°C. A sample volume of 0.1 ml of the super- natant was used in all the determinations.

Uric acid was determined by the method of Liddle et al. (1959) as summarised by Potrikus and Breznak (1980). An assay cuvette contained 2 ml of 0.1 M gly- cine buffer (pH 9.4), 0.1 ml antlion excretory extract and approximately 0.03 enzyme units (U) of purified uricase (hog liver, type V, Sigma). Uric acid (free acid, Sigma) was used as standard.

The allantoin content was determined by the method of Borchers (1977). Allantoin was converted to allantoic acid by dilute alkaline hydrolysis. This was done by heating 0.1 ml antlion excretory extract with 0.25 ml of 0.6 M NaOH at 100°C for 12 min. The allantoic acid was hydrolysed by the addition of 0.5 ml of 0.1% 2,4- dinitrophenylhydrazine (crystalline, Sigma) dissolved in 2 M HCL. Heating was continued for 4 min and the hydrazone of the resulting glyoxylic acid was formed.

The tubes were cooled and alkalified with 2.5 ml 2.5 M NaOH. Absorbance was read at 520 nm after 11 min.

Standard solutions of 50␮M allantoin, 0.83␮M urea and 1.2␮M uric acid were used. The small interference due to uric acid or urea could then be corrected for.

Urea and ammonia were determined with the ultra- violet method described by Anonymous (1980). An assay cuvette consisted of 1 ml of 0.5 M triethanolamine buffer (pH 8.6), 0.1 ml of 6 mM NADH, 1.9 ml distilled water and 0.1 ml antlion excretory extract. The enzymes L-glutamate dehydrogenase (0.2 mg, bovine liver type III, Sigma) and urease (0.05 mg powder, Sigma) were added. Changes in NADH absorption were measured at

(3)

340 nm. Urea (crystalline, Sigma) and ammonium sulphate (grade I, Sigma) were used as standards. An extinction coefficient of 6.3 cm²/␮mole was used (Anonymous, 1980). Samples of a standard solution of ammonium sulphate, dissolved in 0.6% Li2CO3 were either preheated at 100°C for 10 min (three replicates), or kept at room temperature (three replicates) to deter- mine the loss of ammonia during the preparation of the antlion excretory extracts.

2.1.1. Data analysis

The cumulative body weight of termites used during the feeding experiment for fed Cueta sp. and fed F.

intermedia larvae did not differ significantly (Kruskal- Wallis test, N ⫽ 31, P > 0.05), and comparisons were therefore possible. The quantity of dry weight extracted from the termites during feeding was determined as described earlier (Van Zyl et al., 1997). The nitrogen quantity excreted was calculated as the sum of nitrogen present in the quantities of uric acid, allantoin, urea and ammonia excreted.

Statistical tests were applied according to Zar (1984).

The Kruskal-Wallis test and nonparametric Tukey-type multiple comparisons were used to test for differences between the feeding groups. The Mann-Whitney test was used to test for differences between the amount of nitrogenous excretory products within a feeding group.

2.2. Amino acids

The concentrations of the various amino acids in the excreta were determined in a separate experiment. Seven third instar Cueta sp. and 14 F. intermedia larvae were starved for seven days and fed one termite (mean body weight approx. 20 mg), after which excreta was collected for two days. The excreta was homogenised in 0.5 ml of 10% iso-propanol (Merck).

The amino acids in the antlion excretory extract were then derivatised according to Bidlingmeyer et al. (1984) and Cohen et al. (1984) before further analysis. The samples were derivatised as follows. Subsamples (0.25 ml) of the antlion excretory extracts were transferred to 50 mm⫻6 mm tubes. These subsamples were freeze-dried after the addition of 0.03 ml ethanol: water:

triethanolamine (2:2:1, by volume). They were freeze- dried again after the addition of 0.03 ml ethanol: water:

triethanolamine: phenylisothio-cyanate (PITC, protein sequencing grade, Sigma) (7:1:1:1, by volume). After derivatisation, the samples were left at room temperature for 20 min. They were then diluted in 0.16 M sodium acetate buffer (2:1 by volume), dissolved in an ultrasonic bath and centrifuged for 5 min at 4000 g before a sub- sample (1–4␮l) was used in the analysis. The two mobile phases of the high pressure liquid chromatography (HPLC) system consisted of 0.16 M sodium acetate buffer containing 0.05% triethanolamine (99%,

Merck)(pH 6.35) and 61% acetonitrile (ACS reagent, Sigma). Amino acid concentrations of 25 nM (Boehringer Mannheim) were used for standardisation.

3. Results

Allantoin was the main nitrogenous product in the excreta of Cueta sp. and F. intermedia larvae and occurred in significantly larger concentrations (Fig. 1) and quantities (N ⫽ 13–17, P ⬍ 0.05) than ammonia, urea and uric acid in each species. Ammonia was present in similar concentrations (Fig. 1) and quantities (N ⫽ 16–17, P > 0.05) than urea in the excreta of fed Cueta sp. and fed F. intermedia larvae. The concentrations of the standard ammonium solutions preheated at 100°C or kept at room temperature were identical to within measurement uncertainty, indicating that significant amounts of ammonia were not lost during the preparation of the antlion excretory extracts. Uric acid occurred in low concentrations of only 0.35–1.05% of the allantoin concentrations.

Nine amino acids were detected by high pressure liquid chromatography in the excreta of Cueta sp. larvae (Table 1), with lysine and methionine having the highest concentrations. In the excreta of F. intermedia larvae 13 amino acids occurred in relatively small concentrations, with lysine and tyrosine exhibiting the highest concentrations (Table 1). Not all the amino acids were present in the excreta of the individual antlions. For example valine and alanine were present in the excreta of all the F. intermedia larvae (100%, Table 1), while alanine was present in the excreta of all the Cueta sp. larvae, but

Fig. 1. Mean⫾standard deviation of the concentrations of the main nitrogenous excretory products of fed Cueta sp. (n ⫽ 4), fed F.

intermedia (n⫽8) and starved (n⫽8) F. intermedia larvae. Means with different symbols (*#⫹) are significantly different at P⬍0.05 between excretory products and between feeding groups.

(4)

Table 1

Amino acid concentrations (g/l)^†identified in the excreta of Cueta sp. (n⫽7) and F. intermedia (n⫽14) larvae

Amino acid Cueta sp. F. intermedia

Lysine 0.329⫾0.04 (57%) 0.135⫾0.1 (79%)

Methionine 0.319⫾0.11 (57%) 0.030⫾0.009 (14%)

Tyrosine 0.101⫾0.06 (57%) 0.133⫾0.262 (43%)

Phenylalanine 0.017⫾0.01 (43%) 0.036 (7%)

Valine 0.062⫾0.05 (36%) 0.032⫾0.019 (100%)

Serine 0.085⫾0.08 (29%) 0.006 (7%)

Aspartic acid 0 0 0.031⫾0.001 (21%)

Isoleucine 0 0 0.016⫾0.009 (71%)

Glutamic acid 0 0 0.011⫾0.004 (14%)

Threonine 0.009 (14%) 0.006⫾0.003 (43%)

Alanine 0.007⫾0.004 (100%) 0.006⫾0.002 (100%)

Glysine 0 0 0.005⫾0.002 (21%)

Leucine present present

Total^‡ 0.511 0.269

†Mean⫾standard deviation, with in brackets the percentage larvae where the amino acid was detected in the excreta.

‡Summed amino acid concentrations.

the excreta of only 36% of Cueta sp. contained valine (Table 1).

3.1. Excretion during starvation and feeding periods

The comparison between the nitrogenous products excreted during starvation and feeding periods was restricted to F. intermedia, as both starved and fed larvae of this antlion species were used in these experiments.

Starved F. intermedia larvae ingested significantly less dry weight and excreted a significantly smaller volume of urine during the 36 day starvation period compared to that of fed F. intermedia larvae during the 14 day feeding period (Table 2). Starved F. intermedia larvae also produced excreta less often, namely they did not produce excreta on eight days (average) during the first 14 days after feeding. Fed F. intermedia larvae did not produce excreta on five days (average) during the 14 day feeding period. Starved F. intermedia excreted also sig- nificantly smaller quantities of each nitrogenous excret- ory product than fed F. intermedia larvae (N⫽ 17–27, P⬍ 0.05).

The concentrations of each nitrogenous excretory product did not, however, differ significantly between fed

Table 2

The volume of urine produced, dry weight ingested and nitrogen excreted by third instar fed Cueta sp. and fed and starved F. intermedia larvae

Fed Cueta sp. Fed F. intermedia Starved F. intermedia

Experimental period (days) 14 14 36

Dry weight ingested (mg)^† 9.3^b⫾2.9 (9) 9.4^b⫾5.0 (14) 3.4^a⫾1.0 (16)

Nitrogen excreted (mg)^‡ 0.272 0.204 0.080

Volume excreted (␮l) 29.3^b⫾6.8 (9) 17.6^b⫾7.1 (13) 9.0^a⫾3.2 (16)

†Mean⫾standard deviation, with sample size in brackets. Means with different symbols (a,b) are significantly different at the P⬍0.05 signifi- cance level.

‡Accounted nitrogen calculated as the sum of nitrogen present in the quantities of allantoin, uric acid, urea and ammonia excreted.

and starved F. intermedia larvae (Fig. 1). The excreta of starved F. intermedia larvae comprised both the urine excreted directly after feeding one termite as well as the urine excreted during the 36 day starvation period. Com- parisons between the concentrations of the excretory products of starved and fed F. intermedia larvae could therefore not be directly related to excretion during feeding and non-feeding periods.

There was, however, a tendency as shown by the fact that the concentrations of the water soluble excretory products were higher in the excreta of fed larvae i.e.

ammonia was 1.08 times and urea 3.34 times the concentrations of that obtained for starved larvae. The water insoluble products i.e. uric acid concentration was higher for starved larvae i.e. 1.41 times the concentration of that obtained for fed larvae (Fig. 1). The average allantoin concentration in the excreta of starved larvae was 78.5%

of that in fed larvae.

Van Zyl et al. (1997) demonstrated that when third instar F. intermedia larvae were offered one sixth instar H. mossambicus larvae, 44% of the dry weight extracted from the termites consisted of protein. Starved F.

intermedia larvae ingested 3.4 mg dry weight (Table 2) and thus approximately 1.51 mg protein and 0.242 mg

(5)

nitrogen. When the accounted nitrogen excreted is con- sidered (Table 2), starved F. intermedia larvae excreted approx. one third of the nitrogen initially fed to them over the 36 day starvation period.

Sixth instar H. mossambicus larvae consisted 71.8 ⫾ 4.7% (n ⫽ 91) of water and third instar F. intermedia larvae extracted 56–62% of the available water in the prey (Van Zyl et al., 1997). In the present study starved F. intermedia larvae extracted therefore approx. 8.5 mg water from their prey (body weight 19.1 mg). The volume excreted by these antlion larvae (9.0 mg or ␮l, Table 2) over the 36 days was close to the estimated volume of water extracted from their prey.

3.2. Excretion by the pitbuilder and non-pitbuilder

When the excretory products of the fed larvae of the pitbuilder, Cueta sp., and fed larvae of the non-pit- builder, F. intermedia, were compared, no significant differences were detected in the dry weight quantities ingested or the volume of urine produced after feeding (Table 2). No statistically significant differences exist between the concentrations (Fig. 1) or the quantities (N

⫽ 17–23, P > 0.05) of ammonia, allantoin and urea in the urine of fed Cueta sp. and fed F. intermedia larvae (Fig. 1).

4. Discussion

The present study demonstrated that allantoin was the main nitrogenous product in the excreta of Cueta sp. and F. intermedia larvae. The small quantities of uric acid in the excreta were in contrast to the expectation that uric acid is the main nitrogenous excretory product of terrestrial insects (Wigglesworth, 1965; Chapman, 1983). The findings of the present study are also in con- trast to the observations that the antlion larva U. nostras produced ten times more uric acid than allantoin in its excreta (Razet, 1961 quoted by Bursell, 1967). Urea, ammonia and amino acids were not determined in the excreta of U. nostras. A broader comparison of the excretory products of different antlion species may clar- ify this discrepancy between U. nostras and Cueta sp.

and F. intermedia larvae.

As uric acid is less soluble than allantoin, its excretion is usually regarded as an important mechanism to conserve water (Bursell, 1970). It is hypothesised that the pattern of nitrogen excretion, observed in these antlion larvae, can be linked to their discontinuous alimentary canal and the need to eliminate excess water, salts and nitrogen after feeding rather than the conservation of water during starvation. The question arises therefore, which mechanisms are used by Cueta sp. and F. interme- dia larvae to conserve water. This will be explained

below, where we find in the larval cotton stainer, Dys- dercus fasciatus Signoret a basis for comparison.

D. fasciatus also has a discontinuous alimentary canal (Berridge, 1965a) and allantoin is the dominant nitrogenous excretory product with a concentration of 0.7–2.0 g/l in its liquid excreta after feeding (Berridge, 1965b). This is lower than the estimated 21.1–26.9 g/l in the excreta of Cueta sp. and F. intermedia larvae. Some of the allan- toin in the antlion’s excreta was probably not in solution, but was excreted, as a salt as Bursell (1967) reported the solubility of allantoin as 0.6 g/l in water.

Berridge (1965b) argued that allantoin, rather than uric acid, is the dominant excretory product of D. fasci- atus as allantoin is ten times more soluble in water than uric acid (Bursell, 1967). Allantoin, due to its solubility, can attain high concentrations in the haemolymph of D.

fasciatus; this ensures rapid filtration of allantoin into the malpighian tubules. Consequently filtration can occur without the active transport of allantoin across the malpighian tubule membrane. Filtration can also occur without the associated movement of large volumes of water, due to the large allantoin concentration gradient between the haemolymph and the malpighian tubules (Berridge, 1965b). However, D. fasciatus excreted large quantities of fluid due to the inability of the rectum to reabsorb water (Berridge, 1965b). In D. fasciatus these large quantities of fluid are needed to eliminate excess ions, while there is little recycling of water within the insect (Berridge, 1965b).

For Cueta sp. and F. intermedia larvae this argument of Berridge (1965b) may be valid during periods of food abundance when large quantities of fluid must be elimin- ated, especially as Cueta sp. and F. intermedia larvae ingested large quantities of nitrogen i.e. nucleic acids and proteins (Van Zyl et al., 1997). However, during starvation periods the elimination of nitrogen as allantoin rather than uric acid would seem to be disadvantageous.

Under these circumstances urine is retained in the rectal pouch where reabsorption takes place through the modified epithelium of the rectal fold and the six cryp- tonephric malpighian tubules which are laterally dis- placed on the rectal pouch (Van Zyl, 1995).

Wigglesworth (1965) demonstrated this experimentally by injecting dyes into Myrmeleontidae larvae. As fluids entered the perinephric space, some were absorbed by the modified epithelium and returned to the haemocoel.

As the excreta of starved F. intermedia larvae consisted of urine produced directly after feeding as well as during the 36 day starvation period, the present study didn’t demonstrate the effect that reabsorption of water from the rectal pouch would have on the concentrations of the excretory products. The smaller volumes excreted by starved larvae over 36 days compared to the urine pro- duced by fed F. intermedia larvae over 14 days can be in part the result of reabsorption, but in part also the

(6)

result of the smaller quantities of water ingested by starved larvae.

However, during starvation F. intermedia larvae can diminish water loss, not only through adjustments made in excretion, but also by closing the spiracles to mini- mise evaporation. It has been previously reported that a low relative humidity of 22% had no significant influ- ence on the body weight of F. intermedia (Van Zyl, 1995). This is an important feature as low relative humidities (20–65%, Van Zyl, 1995) are characteristic of their natural environment. Youthed (1973) proposed also that Myrmeleon obscurus Rambur and Cueta lance- olatus larvae actively control water loss by closure of the spiracles, as the rate of water loss at 0% relative humidity increased after death in these larvae.

It is suggested that in F. intermedia larvae the water loss through excretion and evaporation during starvation is replaced by water derived from lipid catabolism and through reabsorption from the stored urine. Water derived from lipid catabolism also contributes to the water gain of these larvae, because F. intermedia depends on lipids as an energy resource during starvation (Van Zyl, 1995). It is therefore suggested that the elimination of excess ions, nitrogen and water during periods of food abundance, rather than the conservation of water during starvation, is the critical factor for Cueta sp. and F. intermedia larva. This is probably an important reason why allantoin, and not uric acid, is the dominant nitrogenous excretory product.

The quantities of urea in the excreta of Cueta sp. and F. intermedia can probably be regarded as metabolic by- products, as suggested by Cochran (1985) for other insects. The presence of ammonia is not such an uncom- mon phenomenon, as Cochran (1975) stated that ammonia excretion should pose no problem for terrestrial insects voiding a wet excreta. Ammonia contributes only 3.6% and 4.0% of the nitrogen in the excreta of fed Cueta sp. and fed F. intermedia larvae, respectively.

This in contrast to the aquatic neuropteran larva, S. luta- ria, which can afford to excrete 90% of its nitrogen as ammonia (Staddon, 1955). When water is limited, as in the terrestrial adult stage, S. lutaria excretes predomi- nantly uric acid (Staddon, 1955). The availability of water is thus crucial for ammonia excretion.

The wide range of amino acids in the excreta of Cueta sp. and F. intermedia larvae is not unusual for insects.

For example fifteen amino acids were observed in the excreta of the crane-fly, Tipula paludosa Meigen (Griffiths and Cheshire, 1987), seven in the excreta of Rhodnius prolixus Stal (Wigglesworth, 1965) and six in the excreta of D. fasciatus (Berridge, 1965b). C. carnea, Chrysopa cubana and Chrysopa rufilabris Burmeister eliminated an adhesive substance during the larval stage (Spiegler 1962b) containing␣-amino acids which is an end product of nitrogen metabolism (Spiegler, 1962a).

Finally, as the non-pitbuilder, F. intermedia was twice

as large as the pitbuilder Cueta sp. it can be concluded that neither the pitbuilding and non-pitbuilding lifestyles nor the differences in body weight influence excretion to a large extent. Feeding conditions seem to be the major influence on the excretory products.

Acknowledgements

We thank the Department of Microbiology of the Uni- versity of the Free State for technical assistance and the use of their facilities; J. Grimbeek of the University of Pretoria for his statistical advice, J.D. Mitchell, L.H. van Zyl and F. D. Duncan for comments on earlier drafts and the Department of Zoology and Entomology of the University of Pretoria for the use of their facilities in preparing the manuscript. The Foundation of Research Development provided financial support.

References

Anonymous, 1980. Methods of Enzymatic Food Analysis. Biochemica 6800, Mannheim 31, West Germany.

Berridge, M.J., 1965a. The physiology of excretion in the cotton stainer, Dysdercus fasciatus Signoret I: anatomy, water excretion and osmoregulation. Journal of Experimental Biology 43, 511–521.

Berridge, M.J., 1965b. The physiology of excretion in the cotton stainer, Dysdercus fasciatus Signoret III: nitrogen excretion and excretory metabolism. Journal of Experimental Biology 43, 535–

552.

Bidlingmeyer, B.A., Cohen, S.A., Tarvin, T.L., 1984. Rapid analysis of amino acids using pre-column derivatization. Journal of Chrom- atography 336, 93–104.

Borchers, R., 1977. Allantoin determination. Analytical Biochemistry 79, 612–613.

Bursell, E., 1967. The excretion of nitrogen in insects. Advances in Insect Physiology 4, 33–67.

Bursell, E., 1970. An Introduction to Insect Physiology. Academic Press, New York.

Chapman, R.F., 1983. The Insects: Structure and Function, 3rd ed.

Hodder and Stoughton Ltd., Hong Kong.

Cochran, D.G., 1975. Excretion in insects. In: Candy D.J., Kilby B.A.

(Eds.), Insect Biochemistry and Function. Chapman and Hall Ltd., London, pp. 177–282.

Cochran, D.G., 1985. Nitrogenous excretion. In: Kerkut G.A., Gilbert L.I. (Eds.), Comprehensive Insect Physiology, Biochemistry and Pharmacology. Pergamon Press, Oxford, vol. 4, pp. 467–506.

Cohen, S.A., Tarvin, T.L., Bidlingmeyer, B.A., 1984. Analysis of amino acids using precolumn derivatization with phenylisothiocy- anate. American Laboratory (August).

Gaumont, J., 1976. L’appareil digestif des larves de planipennes, Annales des Sciences Naturelles Zoologie et Biologie Animale 12, 145–250.

Griffiths, B.S., Cheshire, M.V., 1987. Digestion and excretion of nitro- gen and carbohydrate by the crane-fly larva Tipula paludosa (Diptera: Tipulidae). Insect Biochemistry 17, 277–282.

Liddle, L., Seegmiller, J.E., Laster, L., 1959. The enzymatic spectro- photometric method for determination of uric acid. Journal of Lab- oratory and Clinical Medicine 54, 903–913.

Lozı´nski, P., 1911. U¨ ber die Malpighischen Gefa¨sse der Myrmeleoni- denlarven als Spinndru¨sen. Zoologisher Anzeiger 38, 401–417.

(7)

Potrikus, C.J., Breznak, J.A., 1980. Uric acid-degrading bacteria in guts of termites Reticulitermes flavipes (Kollar). Applied and Environmental Microbiology 40, 117–124.

Shaw, J., 1955. Ionic regulation and water balance in the aquatic larva of Sialis lutaria. Journal of Experimental Biology 32, 353–382.

Spiegler, P.E., 1962a. The origin and nature of the adhesive substance in larvae of the genus Chrysopa (Neuroptera: Chrysopidae). Annals of the Entomological Society of America 55, 69–77.

Spiegler, P.E., 1962b. Uric acid and urate storage in the larva of Chry- sopa carnea Stephens (Neuroptera Chrysopidae). Journal of Insect Physiology 8, 127–132.

Staddon, B.W., 1955. The excretion and storage of ammonia by the aquatic larva of Sialis lutaria (Neuroptera). Journal of Experi- mental Biology 32, 84–94.

Van Zyl, A., 1995. Aspects of the ecophysiology of three antlion species (Neuroptera: Myrmeleontidae) in the Kalahari. Ph.D. Thesis, Department of Zoology-Entomology, University of the Free State, Bloemfontein, South Africa.

Van Zyl, A., Van Der Linde, T.C. de K., Van Der Westhuizen, M.C., 1996. Ecological aspects of pitbuilding and non-pitbuilding antlions (Neuroptera: Myrmeleontidae) in the Kalahari. African Ento- mology 4, 143–152.

Van Zyl, A., Van Der Westhuizen, M.C., Van Der Linde, T.C. de K., 1997. Utilisation of prey by antlion larvae (Neuroptera:

Myrmeleontidae) in terms of energy and nutrients. Journal of Insect Physiology 43, 605–613.

Wigglesworth, V.B., 1965. The Principles of Insect Physiology, 6th ed. Methuen and Co. Ltd., London.

Youthed, G.J., 1973. Some adaptations of myrmeleontid (Neuroptera) and rhagionid (Diptera) larvae to life in hot dry sand. Ph.D. Thesis, Department of Zoology and Entomology, Rhodes University, Grah- amstown, South Africa.

Zar, J.H., 1984. Biostatistical Analysis, 2nd ed. Prentice-Hall Inc., New Jersey.