zy zyxwvu zyxwvu zyxwvutsr zyxwvutsr ENTOMOLOGIA SINICA Volume 3, Number 4 , 1996, pp, 329-337 329 POPULATION SEASONALITY AND LARVAL DEVELOPMENT OF LAGRlA HlRTA L. (COLEOPTERA : LAGRIIDAE) Hongzhang Zhou Institute of Zoology, Acudemia Sinica, Beijing 100080, China (Received Apr. 2 5 , 1996; accepted Jun. 1, 1996) Abstract The seasonality of the field population of Lugria hirtu L. is remarkable: adults occur only from the end of June to the beginning of August, whereas larvae exist in the whole year except June. Based on the data of field observations and larval head-width measuring, the author demonstrated that L. hirtu would reproduce in a very short interval in June to August ; eggs hatch and develop (molt) into the 3rd, 4th or 5th (sometimes 6th) instar before winter; after larval overwintering, the beetles complete the larval development after total 8-10 times of molts and pupate in June. The larval development rate is decreased but not completely stopped during winter (the nonparametric median test, P<O. 01 ). This result changes the hitherto opinion of 4 or %star larval development of I,. hirta and will stimulate the studies on its life history. Key words seasonality, larval developmcnt . l,ogriu hirtu 1 INTRODUCTION Studies on the seasonal patterns and life-history adaptations of insects are becoming more and more important for understanding evolutionary biology. A subset of such studies focuses on the seasonal timing of life-history components (Wolda 1988). Temperature? photoperid and huindity are important abiotic factors for insect development and also crucial cues for insects to anticipate seasonal changes (Danks 1987, Tauber et al. 1986 , Zhoii 1992 >. Tn the temperate zone, different geographical regions have quite different natural conditions, i. e, abiotic factors exhibit different seasonal patterns, and the fauna compositions display also a high degree of local speciality. Such phenomenon reveals that adaptation mechanisms are different in insects inhabited in different areas. Such differences can also be seen in life-history adaptation. A variety of different life-history strategies have been evolved to adjust insect life cycle to fit the seasonal cycle. Morever, insect populations of different species often show a variety of seasonality patterns even at * This study was supported by a scholarship from the Chinese Academy of Science, a stipend ( Graduiertenfoerderung ) from "Die Landesregierung des Landes Nordrhein-Westfalen '' and a grant from BMFT , Germany. 330 zyx zyxwvutsr zy zyxwvut ENTOMOLOGIA SINICA Volume 3, Number 4, 1996 one or the same locality. If we want to understand the nature of life-history adaptation of a certain species , a first-of-all step is to study the seasonal pattern or phenology of its field populations. Lagria hirta I,. is the only common lagriid species in Central Europe, while most of its close relatives exist in the tropics and the subtropical regions. This beetles can be found throughout practically the whole of Europe, yet not so frequent towards the north (Postner 1961, Horion 1965) The beetle is known as a forest pest with a strict univoltine life cycle, with phytophagous adults living in summer and saprophagous larvae thriving from autumn throughout winter to next spring (Postner 1961 , Schwenke 1974). Although it is a much common species since long, its biology and phenology were poorly understood , particularly the larval development. An unequivocal answer can not be given to the questions like how many instars intervene between eggs and pupae; some pretended 4 and some 5. The studies hitherto published revealed a great disagreement (Postner 196 1 Stammer 1929). This paper is focused on the seasonal pattern of field population of I,. hzrta. The phenology and the possible larval development were predicted based on the data of head capsule widths of larvae. 2 MATERIALS AND METHODS 2. 1 Collection sites The study was carried out in a regenerated forest ( 7 to 9 years old) found in the area of a brown coal mine near Cologne (5Oo54’N), Germany. The detail description of the site can be found in Topp et al. (1993). 2. 2 Methods of collection Three methods were used in the field collection: Kempson-method (Kempson bowl extractor) pitfall traps (Barber traps) and emergence traps (photo-eclector). They are frequently used by zoologists in studying the population abundance of soil insects. These methods were applied in conjunction with the investigations of other soil animal groups (Topp et al. 1993), and many useful data were thereby obtained as to the phenology of the species investigated. With these methods, the author took samples twice each month throughout the period from 1988 to 1990. For these methods are often used and have somehow been standardized in soil ecological studies, we can find their detail explanations in many books (see Muhlenberg 1989, Dunger and Fiedler 1989) , so it is not necessary to repeat them here again. zyxwvutsrq zyxwv zy Hongzhang Zhou : Population seasonality and larval development of Lagria hirta L. 2. 3 Statistics 33 1 The statistical analyses were carried out by using SPSS/pc+ computer program. In order to analyze the distribution patterns of frequency to head capsule widths, a nonparametric test, Median test, was selected. This is virtual a kind of test for two independent samples. The Two-Sample Median test can be used to determine whether two populations (samples) have the same median. In running the test, operators must give a command in the following format: NPAR TESTS MEDIAN ((value)] =varlist BY variable(value1 value2) where the value in parentheses after MEDIAN is the median to be used for the test. If a value is not specified, the sample median is used. The two values in parentheses following the variable named after BY specify the categories for the grouping variable. The output includes the observed significance level. For the detail and principle of this methods, please see the reference books of SPSS/pc i(Brosius 1989 ) , and Sokal and Rohlf zyxwvuts zyxwvutsr (1981). 3 RESULTS 3. 1 Seasonality of the field population Fig, 1 gives the result of the seasonal changes of larval and adult population. This re- zyxwv sult was based on the data of Kempson-method and emergence traps, but not pitfall traps. 70, :Ind. /m’ 1 n O N D J F M A M J J A S O N D J Months Adulls -Larvae F M A M I J A S 0 Fig. 1 Changes in the population density of I,. hirta at Sophienhoehe near Cologne, Germany. The study was conducted from Oct. 1988 to Oct. 1990. Larvae were sampled by using Kempson-method, and adults with photo-eclector. Undoubtedly for both adults and larvae, especially for the former, the change in population density displayed remarkable seasonal patterns. T h e adult beetles occurred steadily only in a very short interval in summer (from the end of June to the end of August 1 , whereas the larvae could be seen to exist during almost the whole year except zyxw zyx zyxwv ENTOMOLOGIA SINICA Volume 3 , Number 4 , 1996 332 June. The overwintered larvae disappeared in June and the new generation appeared in July. The old and new generations did not overlap , since the larvae of the two generations -old and new-were never seen simultaneously in the same set of samples. Based on this observation, one could suppose that, first, the development from the pupae of the old generation, through the adult maturation (preoviposition time) and egg zyx zyxwvu zyxwv zyxwvu hatching, to the larvae of the following new generation, might require only about 30-40 days (cf. Fig. 1). Second, pupation and adult eclosion were well synchronized to such an extent amongst individuals in the field population that the stretch of adult eclosion was shorter than the time span required for the development mentioned above. 3.2 Number of larval molts In Fig. 2 , the frequencies (vertical axis) of individual larvae with different head widths were plotted to the value of head capsule widths (horizontal axis). The result was very clear (Fig. 2 , B) : the larvae of 1,. hirta might undergo as many as 10 instars in de- velopment. By measuring the head capsules of the larvae which were reared in the laboratory (constantly at 15 C-25 C) but died at different stages, the author got almost the same conclusion (Fig. 2 , A ) . ;i,ll 8 :%I B LS 0 32 10 48 56 61 72 80 fiead-capsule width :1/100 88 96 104 112 120 128 136 mmJ Fig. 2 bstribution patterns of head capsule widths. The diagram shows the possible number of larval instars of L. hzrta. A. larvae from laboratory rearing; B. larvae collected from the field. L,.,,,(lst to 10th instar of larvae). Nevertheless , if the frequency distribution of the field-collected larvae was compared with that of the laboratory larvae, the former revealed greater variations, without distinct separations between the succeeding instars. From young larval stages t o advanced ones, z zyxwvutsrq zyxwv Hongzhang Zhou : Population seasonality and larval development of Lagriu hirta I,. 333 more individuals tended to have intermediate head widths, This was particularly noticeable in stage IJ7(7th instar) or older. This might be attributed to the influence of environmental heterogeneity, for the rearing conditions in the laboratory were much more h o m e geneous than in the field. zyxw 3 . 3 Larval overwintering 30 - 25 20 - 15 10 - 5 0 July~hug I. I ;:zyxwvutsrqpon 1; zyxwvutsrq May 15 10 5 I lL1 II I Fig. 3 Distribution patterns of larval head capsule widths in different months. Only the field-collected larvae were used. From the top downward, the shifting from left to right indicates larval molting. Thus, the active molting occurs only before and after winter, but not during winter. zyx ENTOMOLOGIA SINICA Volume 3, Number 4, 1996 334 The frequency distribution patterns of the head capsule widths, based on the data of the field-caught larvae, were depicted separately according to the months during which the collection was made. The data were separated and every two months was taken as a group, then the total data were parted into 6 groups: July-Aug. , Sep. -0ct. , Nov. Dec. , Jan. -Feb. , Mar. -Apr. , May (no larvae had been found in June in this set of Sam- zy zyxwvutsrq ples). For all 6 groups, the relations of frequency to head width were plotted as in Fig. 3. All 6 groups were compared in pairs and tested with the Two-Sample Median test (Tab. 1>. Table 1 Result of the Two-Sample Median test. T h e different patterns in Fig. 2 were compared amongst the 6 two-month groups. month JuI. -Aug. Sep. -0ct. *** Nov. -Dec. *** n. s. Jan. -Feb. * * * n. s. n. s. Mar. -.4pr. *** * * * *** * * * May *** *** *** *** Sep. -0ct. Nov. -Dec. Jan. -Feb. Mar. -.4pr. * * * * * -P<O.O001; * * -P<O.Ol. In Fig. 3 , from the top downward, the distribution patterns exhibited a shifting-toright tendency. This means that younger larvae with relatively smoller head widths molt- zyxwvu ed into older ones as the time was progressing. As shown in Fig. 3 , the development of larvae was faster before and after winter, with more frequent molts , than just during the winter from Sep. to Feb. Amongst the 3 groups collected during Sep. to Feb. there were no significant difference from each other , but they all differed significantly from the other non-winter groups (Tab. 1, P<O. 01). By paying detailed attention to the patterns of the three winter groups, one would find that they were not identical: however slight the difference was, the larval development was not completely stopped in the winter. This means that the larvae developed nevertheless at a very slow rate. By comparing Fig. 3 with Fig. 2 , the author found that the 3rd, 4th and 5th instars were the very stages that encountered the approach of winter; only a few of the 6th and 7th instars of larvae appeared much later in Nov. -Dec. Thus , the 3rd , 4th and 5th larval instars might be the most suitable stages for overwintering. In the winter the development was so decelerated that only one molt might possibly occur. zyxwvutsrq zyxwv zy zyxw Hongzhang Zhou : Population seasonality and larval development of Lagria hirta I,. 335 3. 4 Predicting the life cycle By now a framework of the life cycle of L. hirta can be outlined , of course, in a very approximate and simplified way, as in Fig. 4. The adult beetles living from June to August would produce eggs which, after hatching, were able to develop and molt into the 3rd, 4th or 5th larval instar by the beginning of the winter. A few of the larvae were so far developed that they might molt into the 6th or even the 7th instar before the spring. After the winter, the larvae would go through the remains of the larval development, and then pupate in June. A full life cycle was completed when the eclosed adults began to reproduce. zyxwvutsrq Fig. 4 Predicted phenology of L. hirta, based on the measurement of larval head capsule widths and the observations in the field. I (adults), E (eggs) L,-,,(lst to loth instar of larvae), P (pupae). 4 DISCUSSION According to the results given above, two life-cycle features of this species were re- markable: one was the long duration spent in larval stages, and the other was that the beetles might punctuate the larval development into as more as 10 instars (the possibility exists at least >.The duration of larval development was long enough to enable larvae to experience the whole seasonal course of a year and endure all kinds of dramatic weather changes in the habitat area. A kind of life cycle like this, with so long a time spent in the larval stages, can never be understood very well in the sense of adaptation, if the details of larval development 336 zyxw zyx zyx zyxwvutsr ENTOMOLOGIA SINICA Volume 3 , Number 4 , 1996 have not been fairly investigated. So, it is spontaneous that my attention was first concentrated on larvae and the number of instars. In this study, an old but simple and frequently used method -measuring the width of head capsules of larvae - was employed with which the author got the first illumination on the matter, the larval development over the seasonal course. Of course, this was not direct observation. An eventual conclusion can be made only when the rearing research has been successfully completed (Zhou 1992). As a forest pest, L. hirta has been studied by many authors (see reviews from Stammer 1929, Posther 1961, Schwenke 1974). From these studies, we can get a basic knowledge of its biology and phenology. However, all these studies hitherto made have not given a correct result for the larval development. According to my studies, the larval development of this species completed only after 7 to 8 instars, or sometimes 10, not merely 4 to 5. This was a remarkable difference. In viewpoints of ecological strategy of adaptation, insect life histories can be categorized to r-or K-type: the former is usually characterized by short life cycles, high reproduction rate, high mortality and unstable populations, whereas the latter by long-living individuals, moderate fecundity, high surviving ability and relatively stable populations. In the real situations, however distinctions can not be so obvious. L. hirta is a species with phytophagous adults and saprophagous larvae; its reproduction rate is higher and its population shows a great degree of fluctuation. These mean that this beetle species is more like r-type in general. Nevertheless, the strict univoltine life cycle and the long individual life as larva are typical K-type characters. For the adaptation nature of L. hirta, this paper could certainly not be the eventual resolution, but rather the new stimulus which will provoke further investigations. Why can the species L. hirta keep existing with such a life cycle? How can this life cycle co-ordinate with the annual cycle? It was just these problems that formed the main subject and the central goal of my studies following this research (Zhou 1992). My hearty thanks are expressed to Prof. Dr. W. Topp (University Cologne, Germany) who kindly supplied the research room in his laboratory and enthusiastically supervised the whole research. I express my warmest gratitude to Prof. Dr. D. Neumann (University Cologne, Germany) and Prof. Dr. Junde Qin(1nstitute of Zoology, Academia Sinica) for their kindly help and enthusiastic supports. Technical assistance by Mrs. Hong Liu is also appreciated. Acknowledgments zyxwv zy Hongzhang Zhou: Population seasonality and larval development of Lagria hirta L. 337 References Brosius, G. 1989 SPSS/pc+ Advanced statistics and tables. McGraw-Hill. Hamburg. Danks, H. V. 1987 Insect dormancy: an ecological perspective. Biological Survey of Canada (Terrestrial Arthropods), National Museum of Natural Sciences, Ottawa. Biological Survey of Cana- zyxwvutsrq zyxwvutsrq zyxwvutsrq zyxwvuts zyxwv zyx No. l. 439 pp. Methoden der Bodenbiologie. Gustav Fischer Verlag. Stuttgart, N. Y. da Monograph series Dunger and Fried 1989 , 432pp. Horion , A. 1965 Faunistik der mitteleuropaeisohen Kaefer. Bd. X. Ueberlingen-Bodensee. Muhlenberg, M. 1989 Freilandoekologie. 2. Auflage. Quelle &. Meyer, UTB. Heidelberg, Wiesbaden. 430 pp. Postner, M. en. 1961 Schadfrass des Wollkaefers Lagria hirta L. (Lagriidae, Coleoptera) an Jungficht- Anz. Schaedlkde. Schwenke, W. 34: 52-54. 1974 Die Forstschaedlinge Europas, Bd. U . (Kaefer). Verlag Paul Parey, Ham- burg and Berlin. Sokal, R. R. and F. J. Rohlf 1981 Biometry, the principles and practice of statistics in biological re- search. Second edn. W. H. Freeman and Company. San Francisco. 859 pp. Stammer, H. J. 1929 Die symbiose der Lagriiden (Coleoptera). Tiere Bd. 15: 1-34. Tauber, M. J. , C. A. Tauber, and S. Masaki 2. f. Morphol. u. Oekol. d. 1986 Seasonal adaptations of insects. Oxford Univ. Press, N.Y. 411pp. Topp , W. , O.Gemesi, C. Gruening, et d. 1993 Forstliche Rekultivierung mit Altwaldboden im Rheinische Braunkohlenrevier. -Die Sukzession der Bodenfauna. 2001. Jb. Syst. 119(4) :505- 533. Wolda, H. Zhou 1988 Insect seasonality: why? , H. Z. Ann. Rev. Ecd. Syst. 19:l-18. 1992 Life-history adaptations to an unpredictable seasonal environment. Studies on two univoltine beetles Othius punctulatus ( Goeze ) ( Staphylinidae ) and Lagria hirta L. (Lagriidae). Dissertation, Koeln. 176pp.