Pedobiologia 40,240-250 (1996) Gustav Fischer Verlag Jena Role of temperature in habitat selection and activity patterns in the ground beetle Angoleus nitidus J. C. Atienza, G. P. Farinós and J. P. ZabaUos Dep. Biologia Animal 1 (Entomología), Fac. Biología, Universidad Complutense de Madrid, 28040 Madrid, Spain. Surnrnary. Habitat selection and adjustment of diurna] activity rhythm are two strategies that ectothermal organisms use in order to avoid unfavourable environmental conditions. Nevertheless, these patterns are variable in time, and differences in physiological requirements or wheather changes can produce different responses in the population. In this papel' we studied in the field during the breeding period of Angoleus ni/idus Dejean 1828 (Coleoptera: Caraboidea: Pterostichidae): a) the habitat selection patterns of males and females and b) the diurna1 activity pattern of both sexes under different temperature conditions, taking account of the different sexual physiological constraints during this periodo Habitat selection by both sexes was quite similar, although females selected more strongly the habitat where the vegetation was the tallest. This result was interpreted as a search for suitable oviposition sites by femates. On fine days, when the temperature at midday was a limiting factor for the activity of beetles, bimodal activity patterns were found, although ma1e and female behaviour was marginally different. On cloudy days the curves obtained were unimodal being a1so different between sexes. Females were more cautious than males even when temperature was not restrictive. These results were interpreted as a ref1ection of the different temperature tolerance existing between males and females due to physiological differences inherent in the breeding periodo Key Words: Angoleus nitidus, diurnal activity, ground beetle, habitat selection, ground temperature, sexual differences Introduction Insects have developed several strategies for avoiding specific physiologically unsuitable conditions caused by abiotic factors of the environment. These strategies are, basically, of three types: morphological, physiological and behavioural (Casey & Hegel 1981, Zachariassen et al. 1987, Rasa 1994, Rutowski et al. 1994, Schmitz 1994). Behaviourally, insects are capable of avoiding unfavourable conditions by using different microenvironments, choosing particular sites appropriate to their thermal balance (Casey 1981). Ambient temperature has an important role in insects' lives since it affects, either directly or indirectly, their development, survival, abundance and reproduction (Szujeki 1987), and its effect is especially important for small terrestrial ecotherms beca use many of theír physiological functions are related to size (Schmid t-Nielsen 1990; Atkinson 1994). Through habitat selection, animals use space in accordance with their needs. Habitat selection is assumed to have a direct and strong impact on fitness (Orians 1991). Individuals of a species select and utilize a habitat, being adapted to its characteristic features (Evans 1983). The area selected offers a far more secure prospect for a long !ife than would a 240 Pedobiologia 40 (1996) 3 random choice (Klopfer & Ganzhorn 1985) and provides refuge sites which avoid risk of predation and adverse environmental conditions. But insects also have the possibility of exploiting temporarily their surroundings, by adjustment oftheir annual, seasonal, circadian 01' daily activity rhythms, when environmental conditions are restrictive. Ground beetles, like other insects, can use these strategies, but with the added difficulty that a lot of them cannot f1y, and thus most of their life cycle is completed on the ground. Thus, abiotic factors that affect the ground, and especially ground surface temperature, innuence directly habitat selection by, and activity patterns of these insects. For this study we have chosen a ground beetle, Angoleus nitidus Dejean 1828 (Coleoptera: Caraboidea: Pterostichidae), typical of naturallittle pools, in the breeding period of its life cycle. The aim of this paper is to study a) habitat selection by A. nitidus and b) its diurnal activity patterns in summer on two types of days: fine days when high temperatures create restrictive conditions for survival, and cloudy days. We expect differences between sexes in habitat selection and activity rhythms, since females are the carriers of the embryos and so their requirements are different. We also expect differences between males and females in their diurnal activity patterns depending on the climatic conditions on each type of day. On cloudy days these differences would be more c1early marked than on fine days, since on fine days temperature is likely to restriet both sexes equally. Materials and Methods Study species Angoleus niridus Dejean 1828 (Coleoptera: Caraboidea: Pterostichidae) is a medium-sized predatory ground beetle (9,5 -10 mm), which lives in the West Mediterranean Region. Although it ranges throughout Spain, Ilaly, Morocco, Algeria and Tunisia (Magistretti 1965), its distribution depends on the presence of natural water pools. This ground beetle needs high levels of moisture (Antoine 1957) and is common in saline ecosystems (Vives & Vives 1978; Zabal!os 1986), but it is absent from sites which are loo saline. Hence, A. nitülus has a very wide distribulion but lhe populations are very localised and isolaled. This species is very vulnerable at presenl since it depends a 101 on smalJ and saline pools of water and these are being utilized more and more for agricultural purposes. Studyarea The research was conducted on the edge of the Pedrezuela marsh (828 m.a.s.l.), located in the province of Madrid (central Spain), in the summer of 1993 coinciding with the reproduction period of this species. At this time the area is mainly covered by annual plants which are grazed by sheep, but in spring, autumn and winter it is totally or partialJy f1ooded. The study area is inc1uded in the Mediterranean region and has a typical continental climalc with exlremely warm and dry summers. In approximately 20 ha of this area five differenl habilats were defined: A) Earthy place: a very dry area without stones or vegetation. B) Marsh plain: the nearest place to water, with scant vegetation. The dominant plant is Verbena offieinalis L. (Verbenaceae). C) Pasture: this area has a wide variety of plants, most of them characteristic of natural grazing systems. D) Thistly place: this sector is dominated by nettles Ur/iea dioiea L. (Urticaceae) and large prickly Compositae, especialJy Cirsium vulgare (Savi) Ten., Ce11laurea calei/rapa L. and Seolymus grandiflorus Desf. E) Stony place: in addition to its high stone densjty, this js the sector with the highest plant species diversity (Table 1). The soil type was similar in al! sectors. Dala eolleelion Data reJating to habitat selection were colJected by means of pitfalJ traps. Eighteen randomly distributed 3 x 2 grids of traps were set out within the five sectors defined earlier. Each grid contained six traps (plaslic jars wilh a mouth diameter of 6.5 cm and 9.5 cm deep), 2 m apart, and they were completely empty to avoid rejection or attraction towards them (Greenslade J964; Baars 1979; Luff J975). Ground 16 Pedobiologia 40 (1996) 3 241 Table lo The frequency of occurrence of plant species within each of the habitats Family Scientific name Urticaceae Urtica dioica Polygonaceae Polygonum sp. Rumex sp. Rumex? sp. CaryophyJlaceae Earth Marsh Pasture Thistly Stony place plain place place (n = 15) (n = 15) (n = 15) (n = 30) (n = 55) 0.06 0.26 0.8 0.73 1.00 0.06 0.96 0.87 0.01 0.69 Spergularia rubra Spergularia sp. 0.86 0.26 0.03 0.30 0.10 Cruciferae Unidentified sp. 0.06 Leguminosae Biserrula pelecinus Trigonella policeratia Medicago lupulina Medicago polymorpha Medicago sp. Trifoliwn resupinalum Trifolium repens Trifolium arvense Trifoliwn sp. I Trifolium sp. 2 Trifolium sp. 3 Trifoliwn sp. 4 Trifolium sp. 5 Trifolium sp. 6 Trifolium sp. 7 Trifolium sp. 8 Ornithopus sp. 0.33 0.03 0.93 0.06 0.06 0.66 1.00 0.06 0.33 0.06 0.20 0.66 0.40 0.03 0.03 0.03 0.06 0.01 0.49 0.03 0.30 0.09 0.01 0.03 0.78 0.41 0.07 0.06 0.09 0.01 0.03 0.01 0.01 Geranium sp. Erodium sp. 0.03 0.09 0.03 Malvaceae Malva sylvestris 0.03 Umbelliferae Unidentified sp. 0.03 Geraniaceae Verbenaceae Verbena officinalis Labiatae Thymus sp. Mentha pulegium Plantaginaceae Plantago Plantago Plantago Plantago Plantago Compositae 242 0.03 0.03 0.06 major lagopus media sp. I sp. 2 Helichrysum sp. Xanthium spinosum Senecio vulgaris Cirsium vulgare Centaurea calcitrapa* Scolymus grandiflorus Crepis sp. Taraxacum sp. Unidentified sp. 1 Unidentified sp. 2 Unidentified sp. 3 Unidentified sp. 4 Pedobiologia 40 (1996) 3 0.93 0.03 0.05 0.06 0.07 0.03 0.13 0.26 0.27 0.01 0.2 0.13 0.18 0.93 0.13 0.12 0.10 0.01 0.33 0.13 0.06 0.06 0.06 0.73 0.23 0.1 0.46 0.06 0.16 0.2 0.03 0.27 0.56 0.03 0.18 0.14 0.09 Table 1. (Continued) Family Scientific name Earth Marsh Pasture Thistly Stony p]ain place place place (n = 15) (n = 15) (n = 15) (n = 30) (n = 55) Gramineae Polypogon marilimus Polypogon viridis Fes/uca myuros Oreochloa? sp. Gauehinia fragilis Hordellln sp. Bromus s/erilis Vulpia unilaleralis? Vulpia sp. 0.33 Unidenlified spp. 1.00 1.00 0.8 0.03 0.06 sp. ] sp.2 sp. 3 sp.4 sp. 5 0.10 023 0.78 001 0.03 0.01 0.16 0.01 0.01 0.03 0.03 0.33 0.06 0.60 0.06 0.10 0.06 0.10 0.05 0.18 0.05 0.01 0.05 * This species was absent from the samples carried out in the thistly place. However it is one of its characteristic species bectlcs were collected every two hours approximately, from 5: 30 am to 7: 30 pm (Greenwich mean time [GMT]; throughout this paper, al! times are reporled as GMT) for eight days and subsequently released far away from the traps (more lhan lO m). With this method 1614 specimens of A. nitidus were col1ected. Dala employed for sludying diurnal activity were obtained from only five ofthe eighleen grids of traps. These traps were slriclly gathered every two hours over ten days, 415 beetles being captured. All of them were sexed using the larger tarsal dilation of the first pair of male legs. This method has often been used fol' other ground beetles (Jeannel 1941 - 1942), and it has been verified by us for A. ni/idus. Two types of day were distinguished: four fine days and six cloudy days. Fine days were sunny at any time and the maximllm mean temperature, reached between 12: 00 pm and 3: 00 pm, was 47.5 oc. On the other hand, the weather on clolldy days was variable wilh sunny and clolldy periods throughollt the day, and lhe maximum mean lemperature was 37.5 oc. The grollnd temperatllre was recorded on lhe surface every 30 minutes to the nearesl 0.1 oC by means of a digital thermometer. Four variables related to vegetation and three related to stones were measured (Table 2). Vegetal cover (erect, creeping or bare ground) can be important from the point of view of thermoregulation of these small ectothermal organisms and of the pl'edation risk, and the number of plant species may Table 2. Results of Principal Components Analysis (PCA) performed to characterize the study area. PCf and PC2 are the f¡rst and second principal components Variables PCl PC2 1. 2. 3. 4. 5. 6. 7. -0.727* 0.883** 0.349 0.896** 0.807** 0.812** 0.678* -0.665 -0.009 0.864** 0.349 -0367 -0.494 -0.577 Bare grollnd Creeping vegetation covel' Erect vegetation cover N umber of plant species Numbel' ofstones: 10-50cm 2 Number ofstones: 50-100cm 2 Number of stones: 100-200 cm 2 Eigenvalue % Explained variance % ClImulative variance 4.007 57.2 57.2 2.025 28.9 86.1 * P < 0.05, ** P < 0.0] 16* Pedobiologia 40 (1996) 3 243 be important for insect species diversity (see Murdoch et al. 1972). Numbers of stones were also considered because they are potential refuges for ground beetles (Thiele 1977) and their prey. Cireles of 0.2 m radius selected at random (using an algorithm to effect random samplings in circular field plots, see Skalski 1987) at most 10 m from each grid centre, were employed for measuring the variables related to vegetation. Five cireles for each grid were utilized, except for those grids whose variances were large because they had a high plant species diversity, were 10 cireles were sampled in order to stabilize the variance (Eberhardt 1976). One cirele of 2 m radius located at the centre of each grid was employed for measuring the variables related to the stones. Three categories of stones were considered on the basis of ground surface area covered: 1 = 10-SOcm 2 , II = SO-100cm 2 and III = 100 - 200 cm 2 Statistical procedures The five habitats considered were continuous in the field. An ordination multivariate method (Principal Components Analysis, PCA; Bhattacharyya 1981) was used to analyze the variables measured in each grid in order to define objective divisions. PCA elassifies the samples, reducing the dimensions of a single group of data by producing a smaller number of abstract variables (linear combinations of the original variables, principal components) (James & McCulloch 1990). By this method, patterns of covariation between the variables could be established, indicating environmental gradients to which these ground beetles were sensitive. The location of this species in each component (measured by the factor scores) was used to characterize its habitat preferences. Before this analysis, the variables expressed as proportions were arcsin-transformed, and the rest were log-lransformed (Zar 1984). Habitat selection was studied by means of univariate analyses using the divisions resulting from the PCA. First we ascerlained whether or not the defined habitats were used by beetles in proportion to their availability, by means of a chi-sqllare goodness-of-fit test (Pearson slatistic). Later we compllled the Savage (1931) selectivity index for measuring habitat preferences by A. nitidus. The Savage selectivity index (Wi) is defined as the proportion of llnits llsed (U¡) divided by proportion of available llnits (p¡). The more W i differs from 1, the higher is the probability of selection. Selection is positive if Wi > I and negative if Wi < 1. The statistical significance of these measurements was tested by comparing the statistic (w i - 0 2 /se(wY with the corresponding critical value of a chi-sqllare distriblltiol1 with one degree of f¡'eedom, where se(w¡) is the standard error of W i (Manly et al. 1993). We estimated se(w i) on the assumption that there was no selection, so that the standard error of W i was approximated by Pi)/(U+ . Pi)' where ll+ is the nllmber of grollnd beetles captured in each grid of traps, and Pi is the proportion of grids in the habitat i. A maximum 1 % of type 1 error was accepted. Significance level was obtained after applying the Bonferroni correction (Rice, 1989). V(l - Results Results oI the peA The results of the PCA defined two gradients in relation to the landscape physiognomy of the habitats. These gradients were used for locating the eighteen sampling units on the factorial plane (Fig. l). Two principal components with eigenvalue > 1, PC1 and PC2, which together accounted for 86.1 % of the total variance (Table 2), were generated by the analysis. PC1 explained 57.5% of the variance and basically separated bare graund areas (on the negative axis) fram vegetal cover whether creeping or erect (associated with the positive axis). The three variables relating to numbers of stones were also associated, this time with the positive axis. PC1 therefore, described a gradient ofvegetational complexity and stoniness which together can be considered as a gradient of habitat structural complexity. PC2 explained 28.9% of the variance. This factor describes two gradients. One shows a gradient of vegetation structural complexity, since it separates erect vegetation cover and number of species (on the positive axis) fram creeping vegetation cover and bare graund (on the positive axis). Moreover, it describes an inverse gradient of stoniness, as variables relating to numbers of stones are grauped on the negative axis. 244 Pedobiologia 40 (1996) 3 - - - - - - _ . _ - _ _._------.. 2 ｾ >< Q) ｾ o o. E e ¡;; O O '5 i5 .a ｾ N tí 0 2 Q,. ｾ 1iS e E z Fig. 1. Location of samples (n = J8) on the factorial plane defined by the two principal components and representation of the number of Angoleus nitidus collected in each grid of traps (open squares). NlImbers in the graphic indicate the mean nllmber of specimeos col1ected in every sector per grid of traps and day. A: Earthy place; B: Marsh plain; C: Pasture; D: Thistly place; E: Stony place (.) 0 O § Q) ｾ O> -2 -2 ­, 00­50 O pe 1 050­100 0100­200 D >200 Habltat Slruclural Complexlly The 18 sampling units were represented on the plane defined by the two factors, and an approximate division into sectors corresponding with the five habitats previously defined was made, so as to understand the relationship between the sectors and the defined gradients (Fig. 1). An explanatory graphic is presented in order to facilitate the interpretation of these gradients (Fig. 2). Habitat seleclion Angoleus nilidus was present in every grid of traps, but it did not choose every habitat with the frequency predicted by chance (x 2 = 1290, df = 4, P < 0.001). Moreover males and females separately were not distributed as predicted by chance (X 2 = 903 and X2 = 436, df = 4, P < 0.001, respectively). It can be seen from Tables 3 and 4 that the marsh plain and the thistly place were positively selected in every case, and the earthy place, pasture and stony place were negatively selected. On the whole marsh plain was the most strongly chosen habitat (Table 3). Despite the fact that males and females chose marsh plain and thistly place positively, both habitats were selected with a different intensity (X 2 = 11.92; df = 4; P < 0.5). Females selected thistly place more strongly than males, Fig. 2. Explanatory graphic for interpretil1g the gradients defíned by tbe two factors obtained in the PCA. A: Earthy place; B: Marsh plain; C: Pasture; D: Thistly place; E: Stony place PedobioJogia 40 (1996) 3 245 Table 3. Selection indices for the five habitats used by Angoleus nitidus population Habitats Earthy place Marsh plain Pasture Thistly place Stony place Total Number of beetles 72 584 71 527 360 1614 2 Available proportion Selection index X 0.045 0.362 0.044 0.326 0.223 1.000 0.267 3.256 0.395 1.469 0.573 5.960 173.125 1027.274 73.623 101.576 186.78 Table 4. Selection indices for the five habitats used by males and females of Angoleus nitidus Habitats 2 Number of beetles Available proportion Selection index X Males Earthy place Marsh plain Pasture Thistly place Stony place 33 371 36 318 218 0.034 0.380 0.037 0.326 0.223 0.202 3.421 0.332 1.466 0.574 124.033 715.134 54.444 60.604 112.524 Females Earthy place Marsh plain Pasture Thistly place Stony place 39 213 35 222 142 0.061 0.334 0.055 0.327 0.223 0.367 3.005 0.493 1.474 0.501 51.164 320.501 20.440 40.979 127.200 wbereas males chose marsh plain more strongly than females. Fig. 1 sbows the distribution and relative abundance of A. nitidus represented on the plane defined by tbe two factors of PCA. Diurnal activity In the study site Angoleus nitidus was active in tbe daytime. Diurnal acttvtty patterns are represented by means of two curves (Fig. 3): A, on fine days and B, on cloudy days. Tbere were no significant differences between numbers of A. nitidus active on cloudy and fine days (N.S.; U = 28; P = 0.70), but there were significative differences in tbe patterns of activity throughout the day (X 2 = 58.28; df = 6; P ｾ 0.01). On fine days, tbe number of beetles was lower at tbe beginning and at tbe end of the day. At tbe start of the day, activity of A. nitidus increased witb the temperature. When temperature attained 40 oC activity was drastically reduced. Wben temperature fell under 40 oC activity started again (Fig. 3a). On tbese days there were also marginally significant differences in male and female behaviour (Wilcoxon t = 3; P = 0.051). Tbe behaviour of both sexes was very alike, but females reacbed tbe second activity maximum before males (17: 30 an'd 19: 30, respectively). On cloudy days, tbe activity of A. nitidus increased witb time according to temperature. Wben the temperature dropped, activity was also drastically reduced. On tbese days tbere were marginally significant differences in male and female bebaviour (Wilcoxon t = 3; P = 0.051), with males reacbing their activity maximum before females did (15: 30 and 17: 30, respectively). 246 Pedobiologia 40 (1996) 3 10,­­­­­­­­­­­­­­­­­­­­­­, rJ) 8 40 6 30 4 20 2 10 al ｾ <\l ..o 'O ｾ ... 3 o LL­'­'­­­­'­­­­'­'­­­'­9­­':3­0­'­­­­'­­­'1­1'­­­:30­'­­­­'­­­'­­­13­'­:3....l.0­­'­­'­....l.:3­­'O'­'­­­'­1­­'7:­30­'­­­­'­­­'1­ 15 9 30 .LJ: 10,­­­­­­­­­­­­­­­­­­­­­­, E E ro éJ ..... e ro Cloudy days 8 40 4 20 2 10 9:30 "O <1> 11:30 13:30 15:30 17:30 19:30 time of day Fig.3. Diurnal activity patterns of males (bold line) and femaJes (fine line) in reJation to the temperature on two types of days (open circJes) Discussion Because Angoleus ni/idus was in its breeding period in the course of the study, we analyzed habitat selection data separately for males and femates to find differences in selection between sexes (Thomas & Taylor 1990). Analyzing the data, whether for the population as a whole or for the two sexes separately, we observed that this species never remained neutral: al! habitats were positively or negatively selected (Tables 3 and 4). This fact make us think that A. ni/idus is an exigent ground beetle in selecting its habitat. Both males and females showed a strong negative selection of the earthy place, which is very dry, lacks vegetation and makes behavioural thermoregulation difficult for beetles, and both selected the thistly place and marsh plain. However, selection intensity of these habitats was different for the two sexes. Females selected the thistly place more strongly than males, whereas males chose the marsh plain more strongly than females. The stronger selection of the thistly place by females may be due to active searching for oviposition sites suitable to the needs of eggs and young larvae (May 1979; Willmer 1982; Michiels & Dhondt 1990; Sowig 1995). Hence, females showed greater thermoregulatory caution in choosing the thistly place more strongly than the marsh plain, since the former had taller vegetation and more stones and provided far more secure sites. Males, on the contrary, chose more strongly the marsh plain, the nearest place to water where the moisture content was greater, and this selection is more in accordance with the usual habits of this species (Antoine 1957). A. nitidus exhibited activity patterns which differed markedly between sunny and cloudy days (Fig. 3) probably because ectothermy makes ground beetle activity largely dependent on weather. On fine days, ground surface temperature often exceeded 40 oC, the critical upper temperature for ground beetles (Lindroth 1992; Thiele 1977). Hence they had to Pedobiologia 40 (1996) 3 247 take refuge in midday hours, remaining relatively motionless at that time. This adjustment of activity patterns is a behavioural mean often used by both vertebrate and invertebrate ectothermal species to avoid high critical temperatures (Casey 1976, Avery 1978, Andrews & Kenney 1990, Bardoloi & Hazarika 1994, Quiring 1994). In insects, in particular, this strategy is widespread in those that are ground dwellers (Casey 1981). By contrast, on cloudy days the lack of temperature high enough for starting activity may be a limiting factor. Such behaviour also occurs in other insects, as shown by Bardoloi & Hazarika (1994) for Lepidoptera larvae, which remained immobile during the cooler hours of the day until the air temperature increased with a corresponding increase in body temperature. This activity pattern displayed on cloudy days is similar to that obtained by other authors (Luff 1978, Alderweireldt & Desender 1990). On the other hand, we wish to emphasize that diurnal activity patterns were different depending on temperature. This fact helps to refute the idea that ground beetles have definite diurnal activity patterns, as Greenslade (1963) reported for several species. His results are probably due to the fact that the experiments were not carried out under natural conditions. Our results show that activity patterns depend more on external factors (on temperature in this case), than on autonomous rhythmicity. In this sense, Thiele & Weber (1968) concluded that activity rhythms were primarily influenced by Jight­dark cycles and not by autonomous rhythmic periodicity, and later Luff (1978) obtained similar results with Hmpalus rL!fipes (Degeer), a ground beetle which showed phase differences in its activity pattern throughout the catching period each year. Moreover, this author obtained similar results to ours with Notiophilus biguttatus (F.), which had a bimodal activity pattern decreasing at mid­day during hot dry weather in summer; but in cooler weather, mid­day activity started again changing the curve to a unimodal pattern. It is remarkable that females reached their activity maximum at 17: 30, regardless of atmospheric conditions, when temperature fel! below 40 oC, whereas males did not reach this maximum at the same time in terms of type of day (Fig. 3). We think that this strategy used by females may be because A. nitidus is in the breeding period of its Jife cycle, and high temperatures in the study area could be dangerous for embryos. If females were active before this time on changeable days, they risked a sudden temperature rise that would be rapidly reflected in their body temperature. So this pattern would be a means of avoiding dehydration and safeguarding embryos. We think that both habitat selection and activity pattern adjustment of this ground beetle serve the same purpose, namely the choice of physiologically suitable sites which aJlow individuals to survive in adverse conditions, in accordance with the particular needs of each stage and sex at each moment of its Jife cycle. Acknowledgements We wish to thank the following people for their help and contributions to this work: S. P. Farinós, D. García­Cuenca, C. Garriga, J. C. Illera, P. Ingelmo and C. Navas for help in coJlecting data; P. González and J. 1. Gutiérrez for determine the plants. Thanks are extended to two anonymous referees for providing substantial editorial advice on an earJier drafl of this manuscript. We greally appreciate the use of facilities provided by Canal de Isabel II during our field work. This study was partially supported by the Universidad Complutense de Madrid (Project nOPR 189/92-4024). References Alderweireldt, M., Desender, K. (1990) Variation of carabid diel activity patterns in pastures and cultivated fields. In: N. E. Stork (ed.) The role of ground beetles in ecological and environmental studies. Intercept. Andover, Hampshire. Andrews, R. M., Kenney, B. S. (1990) Diel patterns of activity and of selected ambient temperature of the sand­swimming lizard Sphenops sepsoides (Reptilia: Scincidae). Isr. 1. Zoo1. 37,65­73. 248 .­­­­ _ .. Pedobiologia 40 (1996) 3 Antoine, M. (1957) Coléopteres carabiques du Maroc. II partie. Mém. Sc. nat. phys. Maroc. Atkinson, D. (1994) Temperature and organism size ­ A biologicallaw for ectotherms? Adv. Eco!. Res. 25, 1­ 58. Avery, R. A. (1978) Activity patterns, thermoregulation and food consumption in two sympatric lizard species (Podareis muralis and P. sieula) from Central ltaly. J. Anim. Eco!. 47, 143 ­ j 58. Baars, M. A. (1979) Catches in pitfall traps in reJation to mean densities of cara bid beetles. Oecologia 41,25-46. Bardoloi, S., Hazarika, L. K. (1994) Body temperature and thermoregulation of Antheraea assama larvae. Entomo!. exp. app!. 72, 207­217. Bhattacharyya, H. (1981) Theory and methods of factor analysis and principal components. In: Capen D. E. (ed.) The use of multivariate statistics in studies of wildlife habitat. USDA Forest Service. Vermont. Casey, T. M. (1976) Activity patterns, body temperature and thermal ecology in two deserl calerpillars (Lcpidoptera: Sphingidae). Ecology 57, 485 ­ 497. Casey, T. M. (1981) Behavioral mechanisms of thermoregulation. In: Heinrich, B. (ed.) Insect thermoregulation. Wiley & Sons, New York. Casey, T. M., Hegel, J. R. (1981) Caterpillar setae: insulation for an ectotherm. Science 214, 113 J ­ 1133. Eberhardl, L. L. (1976) Quantitative ecology and impact assessment. J. Environ. Manage. 4, 27 ­70. Evans, W. G. (1983) Habitat selection in the Carabidae. Coleopts. Bul!. 37,164­167. Greenslade, P. J. M. (1963) Daily rhylhms of locomotor activity in sorne Carabidae (Coleoptera). Entorno!. exp. app!. 6, 171 ­180. Greenslade, P. J. M. (1964) Pitfall lrapping as a melhod for sludying populations of Carabidae (Coleoptera). J. Anim. Eco!. 33, 301­310. James, F. e.; McCulloch, Ch. E. (1990) Multivariate analysis in ecology and systematics: panacea or Pandora's box? Ann. Rev. Eco!. SYSI. 21, 129­166. Jeannel, R. (1941­1942) ｃｯｬ￩ ｰｬ ｾｲ･ｳ Carabiques. Faune de France. Vol. 39, 40. Lechevalier, Paris. Klopfer, P. H., Ganzhorn, J. U. (1985) Habital selection: behavioral aspects. In: Cody, M. L. (ed.) Habital selection in birds. Academic Press, Inc. Orlando. Lindroth, e. H. (1992) Ground beetles (Carabidae) of Fennoscandia. A zoogeographic study. Part Ir. General analysis with a discllssion on biogeographic principies. Smithsonian Institution Libraries and lhe National Science Foundation. Washington, D.e. Luff, M. L. (1975) Some features inOuencing the efficiency of pilfall traps. Oecologia 19, 345­357. Luff, M. L. (1978) Diel activity patterns of some field carabidae. Eco!. Entorno!. 3, 53­62. Magistretti, M. (1965) Coleoptera Cicindelidae, Carabidae. Fauna d'ltalia vol. III. Ed. Calderini, Bologna. Manly, B. F. J., McDonald, L. L., Thomas, D. L. (1993) Resource selection by animals. Statistical design and analysis for field studies. Chapman & Hal!. London. May, M. L. (1979) Insect thermoregulalion. Ann. Rev. Entorno!' 24, 313 ­ 349. Michiels, N. K., Dhondt, A. A. (1990) Costs and beneflts associated with oviposition site selection in the dragonOy Sympetrum danae (Odonata: Libellulidae). Anim. Behav. 40, 668 ­ 678. Murdoch, W. W., Evans, F. e., Peterson, C. H. (1972) Diversity and pattern in plants and insects. Ecology 53,819­829. Orians, G. (1991) Habitat Selection. Am. Nat. 137 (Supplement), 1­130. Quiring, D. T. (1994) Dicl activity pallern of a nocturnal moth, Zeiraphera eanadensis, in nature. Entomo!. exp. app!. 73, 111­120. Rasa, O. A. E. (1994) Behavioural adaplations to moisture as an environmenlal conslraint in a nocturnal burrow­inhabiling Kalahari detritivore Parastizopus arma/ieeps Peringuey (Coleoplera: Tenebrionidae). Koedoe 37, 57 ­ 66. Rice, W. R. (1989) Analyzing la bies of stalistical tesIs. Evolution 43, 223 ­ 225. Rutowski, R. L., Demlong, M. J., Leffingwell, T. (1994) Behavioural lhermoregulalion al male encounler siles by male butterOies (Asteroeampa leiha, Nymphalidae). Anim. Behav. 48, 833­841. Savage, R. E. (1931) The relation between lhe feeding of lhe herring off the east coasl of England and the planklon of the surrounding waters. Fishery lnvesligation, Ministry of Agriculture, Food and Fisheries, Series 2, 12, 1­88. Schmidt­Nielsen, K. (1990) Animal physiology: Adaplalion and environment. Cambridge University Press. Cambridge. Schmitz, H. (1994) Thermal characterizalion of bullerny wings­ 1. Absorption in relation to different color, surface struclure and basking type. J. therm. Bio!. 19,403­412. Skalski, J. R. (1987) Selecting a random sample of points in circular field plots. Ecology 68, 749. Pedobiologia 40 (1996) 3 249 Sowig, P. (1995) Habitat selection and offspring survival rate in three paracoprid dung beetles: the inf1uence of soil type and soil moisture. Ecography 18, 147 ­ 154. Szujecki, A. (1987) Ecology of forest insects. Dr. W. Junk Publishers. PWN­Polish Scientific Publishers. Warszawa. Thiele, H.­U. (1977) Cara bid beetles in their environments. A study on habitat selection by adaptatiol1s in physiology and behaviour. Springer­VerIngo BerIin. Thiele, H.­U., Weber, F. (1968) Tagesrhythmen der Aktivitat bei Carabiden. Oecologia 1, 315 ­ 355. Thomas, D. L., Taylor, E. J. (1990) Study design and tests for comparing resource use and availability. J. Wildl. Manage. 54, 322­ 330. Vives, J., Vives, E. (1978) Coleópteros halófilos de los Monegros. Boln. Asoc. esp. Ent. 2, 205­214. Willmer, P. G. (1982) Microclimate and the environmental physiology of insects. Adv. Ins. Physiol. 16,1­57. Zaballos, J. P. (1986) Primera contribución al conocimiento de los carábidos (Coleoptera) de las lagunas salinas y subsalinas de la Meseta Norte. Actas de las VIII Jornadas de la Asociación española de Entomologia, Sevilla. pp. 700­709. Zachariassen, K. E., Andersen, J., Maloiy, G. M. O., Kamau, J. M. Z. (1987) Transpiratory water loss and metabolism of beetles from arid areas in East Africa. Comp. Biochem. Physiol. 86A, 403­408. Zar, J. H. (1984) Biostatistical analysis, 2nd. ed. Prentice­Hall. New Jersey. 250 Pedobiologia 40 (1996) 3