Read ASEF 3-2009.indb text version

Ann. soc. entomol. Fr. (n.s.), 2009, 45 (3) : 377-392

ARTICLE

Forensic entomology: a new hypothesis for the chronological succession pattern of necrophagous insect on human corpses

Fabrice Lefebvre & Emmanuel Gaudry *

Département Entomologie, Institut de Recherche Criminelle de la Gendarmerie Nationale, 1 boulevard Théophile Sueur, F-93111 Rosny-sous-Bois Cedex, France * Corresponding author

Abstract. Forensic entomology can help to estimate the time elapsed since death, by studying the necrophagous species collected on a cadaver and its surroundings. The determination of the socalled post mortem interval (PMI or period of first oviposition) is based on the development time of necrophagous dipterans and on the chronological pattern of insects' succession on the corpse throughout the decaying process. In the present study, authors investigated this succession by the analysis of the database of the Department of Forensic Entomology of the French Gendarmerie over 12 years (1992­2003) in order to propose a new approach and a new hypothsesis of the dynamic of necrophagous insects' populations present on a human cadaver over time. For all treated cases, the presence/absence and the oldest development stage of the species were recorded. Data were analyzed by statistical and Wagner parsimony methods. The statistic results allowed the association of groups of species with typical developmental stages. The Wagner parsimony analysis showed that the dynamic of the necrophagous insect population present on a cadaver could be characterised by specific species. Together, they allowed establishing hypotheses of succession of necrophagous insects on human cadavers over time that could be usefull in the PMI estimation. Résumé. Entomologie forensique : une nouvelle hypothèse de succession chronologique d'insectes nécrophages sur les cadavres humains. L'entomologie légale permet d'estimer le temps écoulé depuis la mort grâce à l'étude des insectes nécrophages collectés sur un cadavre et dans son environnement. La détermination du délai post-mortem (Post Mortem Interval ou PMI) ou de la période de première oviposition (ponte) est basée sur l'étude des temps de développement des Diptères nécrophages et de la succession des insectes sur le corps en fonction des états de décomposition. Dans cette étude, nous analysons la succession des insectes nécrophages provenant de la base de données du Département d'Entomologie de l'Institut de Recherche Criminelle de la Gendarmerie Nationale (IRCGN) au travers de 12 années d'activité (1992­2003). Le but de cette étude est de proposer une nouvelle approche et une nouvelle hypothèse de la dynamique des populations d'insectes nécrophages présents sur un cadavre au cours du temps. La présence/absence et le stade de développement le plus ancien par espèce sont recensés. Les données sont analysées statistiquement et par la méthode de la parcimonie de Wagner. L'analyse statistique met en évidence des associations d'espèces en fonction de stades types de développement. La parcimonie de Wagner montre que la dynamique des populations d'insectes nécrophages présents sur un cadavre, peut être caractérisée par la présence de certaines espèces. Elle permet également d'établir des hypothèses sur la succession des insectes sur les cadavres humains au cours du temps, pouvant être utilisées dans l'estimation de délais post-mortem.

Keywords: Forensic entomology, necrophagous species, development stage, succession, post mortem interval.

he main aim of the forensic entomology is to estimate a post-mortem interval (PMI) after the discovery of a cadaver. This estimation is mainly based on the statement that the first ovipositions of necrophagous insects on the corpse occur at a period close to the death. When a corpse colonized by necrophagous insects is found, two situations could be considered. The specimens growing on the corpse are identified either

T

E-mail: [email protected] Accepté le 28 mai 2009

as pioneer or later necrophagous species in the process of colonization. In the first situation (pioneer species), the age of the oldest specimens is estimated in order to determine a minimum post mortem interval (Amendt et al. 2007). This determination of the time of death is often used in forensic entomology. In the second situation, later species colonize the cadaver with a delay (i. e. after the pioneers). As it is not possible to deduce directly the period of death, the estimation of the post-mortem interval was only possible by the determination of the chronological succession in the insect colonization process. It is called maximum PMI (Amendt et al. 2007).

377

F. Lefebvre & E. Gaudry

Table 1. Listing of expertise works from 1992 to 2003.

Year Number of cases 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 6 14 36 29 31 30 22 36 35 35 43 39

The first studies establishing a chronological succession pattern of necrophagous insects on a "corpse" (animal carcass or human cadaver) were published at the end of the XIXth century (Mégnin 1894; Johnston & Villeneuve 1897; Motter 1898; Yovanovitch 1888). They mainly reported a chronological succession of ne-

crophagous insects on the cadaver, organised in eight waves and theorized by Mégnin (1894). Later, several entomologists completed similar experiments (Leclercq 1978; Smith 1986; Schloenly et al. 1992; Anderson 2001). These articles completed the Mégnin's results describing chronological colonization on corpses by the local necrophagous fauna depending on the season and geographic area (Anderson 2001). However, the studies were usually based upon single cases and no hypothesis dealing with the necrophagous population had been tested. The purpose of this study is to establish scientifically the chronological succession pattern of necrophagous

Table 2. Listing of cadaver discoveries by year and environmental conditions associated. Buried a 0 2003 38 2002 42 2001 34 2000 35 1999 34 1998 21 1997 30 1996 29 1995 28 1994 33 1993 14 1992 6 Absent: 0 / Present : 1 Biotope e 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994 1993 1992 0 18 16 10 12 11 11 10 12 12 10 5 3 1 7 20 19 17 18 5 13 17 10 14 6 3 2 14 7 6 6 7 6 7 2 7 12 3 0 1 1 1 1 0 2 1 0 2 1 3 0 0 Packaged b 0 2003 30 2002 42 2001 31 2000 32 1999 32 1998 18 1997 28 1996 28 1995 29 1994 35 1993 12 6 1992 Absent: 0 / Present: 1 1 9 1 4 3 4 4 2 3 0 1 1 2 Humidity c 0 2003 36 2002 40 2001 32 2000 32 1999 34 1998 18 1997 28 1996 27 1995 27 1994 32 1993 14 6 1992 Absent: 0 / Present: 1 1 3 3 3 3 2 4 2 4 2 4 0 0 Environment d 0 1 2 3 2003 23 6 9 1 2002 31 5 7 0 2001 25 8 2 0 2000 27 8 0 0 1999 30 3 3 0 1998 14 1 6 1 1997 21 3 4 2 1996 26 2 3 0 1995 21 5 3 0 1994 20 9 7 0 1993 10 2 2 0 6 0 0 0 1992 Outdoor: 0 / Indoor: 1 / Shelter: 2 / Waste plane: 3 Season g 0 19 17 18 18 12 13 10 10 14 20 5 4 1 11 11 6 10 8 2 10 6 7 2 5 1 2 1 1 1 0 1 0 1 2 0 3 0 0 3 7 13 7 7 14 7 6 12 8 9 3 1 4 1 1 3 0 1 0 3 1 0 2 1 2 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994 1993 1992 0 34 33 31 22 27 15 24 23 23 28 12 5 1 5 10 4 15 9 7 6 8 6 8 1 2

Climatic area f 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994 1993 1992

Open rural: 0 / Closed rural: 1 / Urban: 2

a b

Atlantic: 0 / Continental: 1 / Alpine: 2 / Mediterranean: 3 / Tropical: 4

Favourable: 0 / Unfavourable: 1

Buried: human cadaver fully buried at discovery. Packed: human cadaver packed in a plastic bag, a covert, .... c Humidity: presence or absence water in the surrounding of the cadaver. d Environment discovery - habitation: flat, house - shelter : hut, depot, caravan, cave, car, , box, dustbin, ... e Characteristics of the biota (death scene). f Characteristics of climatic areas (death scene). g Season favourable or unfavourable to the insect activity at death scene.

378

Necrophagous insects succession

insects of forensic importance on cadavers. The authors exploited the database of the Forensic Entomology Department of the Forensic Science Institute (French Gendarmerie), gathering information from legal investigations of human cadaver discoveries and PMI estimated. This study proposed to test the theory of Mégnin, namely that there is a chronological succession of necrophagous insects on the cadaver, organised in eight waves (Mégnin 1894).

Material and Methods Material The study was based on 356 expertise works performed by the Forensic Entomology Department between 1992 and 2003 (tab. 1) on the entire French territory (mainland and overseas territories). Collection of insects was mainly performed by crime scene technicians, in respect of a defined sampling protocol (Amendt et al. 2007). Several parameters of these legal expertises were recorded in a database: species inventory, development stage of specimens, environmental and weather conditions, location, biotope, PMI estimation and also feedback on results (confirmed cases) - (tab. 2). We only took into account in this study the presence/absence of taxa, the development stage of specimens and the PMI estimation. The PMI (estimated and confirmed by feedback) were classified in six homogeneous classes and listed in tab. 3. Over the 356 expertises, 86 cases were solved by criminal investigations (confirmed cases). Requests are annually sent regarding cases treated at year n­4 to obtain feedback on results. This delay is induced by the time required by the judicial process for criminal cases. The listing of taxa preserved in all cases (x 356) was performed under specific conditions. For this study, specimens sampled in cadavers were not considered with the same importance, according to the species and their development stage. Coleoptera specimens were thus considered either as adult, immature (larvae) or post-immature development stages (exuviae). In contrast, Diptera, Hymenoptera and Lepidoptera specimens were only considered as immature (larvae, pupae, nymph) or post-immature stages (puparia, exuviae). These stages are directly linked to the cadaver that constitutes a nutritional substrate. Additionally, only the oldest specimen from each species, characteristic of the longer time period spent on the cadaver, was considered (Amendt et al. 2007). We took in account in all analyses only insects species present at least two times among the data sets. All other specimens of Arthropods (tab. 4) were not considered in the present study. Finally, 117 taxa were identified in the sampling (tabs. 5­8). Several taxa could simultaneously be observed on a corpse. Out of 356 cases, the Diptera order was predominant. Indeed, in the final inventory, Dipterans were present in 99.9 % of cases (353 cases out of 356). The frequency of Calliphoridae was respectively 93 % (331 cases) and that of the Muscidae was 45% (158 occurences). The Coleoptera, gathering several necrophagous species of forensic importance, were identified 191 times (50 %). The Hymenoptera and Lepidoptera were respectively observed 26 and 12 times (7.3 % and 3.4 %). The main objective of the study is to determine the succession of the different `waves' on the corpse related to time. To establish

Table 3. PMI estimation listed by classes. Post Mortem Interval (PMI) Undetermined PMI < 1 week 1 week < PMI < 2 weeks 2 weeks < PMI < 1 month 1 month < PMI < 3 months PMI > 3 months

a

Number of cases 4 91 65 74 76 46

% 1.10 25.60 18.30 20.80 21.30 12.90

Confirmed cases 1a 21 17 15 20 13

% 1.10 24.20 19.50 17.30 23.00 14.90

case not considered in Wagner parsimony analysis

the presence of a succession of functional ecological groups of insects on a cadaver, a single species cannot be considered independentely. A taxonomic identification of genus or family could theoretically be characteristic of its functional group. For example, in the genus Lucilia sp. Robineau-Desvoidy 1830 (Calliphoridae), it is irrelevant to record the number of cases where Lucilia sericata (Meigen 1826), Lucilia caesar (L. 1758), Lucilia illustris (Meigen 1826) or Lucilia ampullacea Villeneuve 1922 occurred, because these are four different taxa with similar biology and considered with the same ecological function in forensic entomology (Smith 1986). If the species level was considered, four taxa should be analysed. Thus, only one genus gathering these different species was adequate for the study. This situation is rather particular in synecology, as Nel et al. (1998) reported that in freshwater biomonitoring for

Table 4. Inventory of taxa from other sampled groups. Order Collembola 1 Dermaptera 1 Hemiptera 1 Orthoptera 1 Mecoptera 1 Psocoptera 1 Thysanoptera 1 Chilopoda Family Diet Totala 2 Saprophytophagous 3 Saprophagous ­ Predaceous 6 Phytophagous 2 Phytophagous 3 Predaceous 1 Mycophagous 1 Phytophagous 6 Predaceous 8 Predaceous 2 Predaceous 2 Saprophytophagous 1 Predaceous 11 Predaceous 3 Predaceous 2 Predaceous 18 Saprophagous 3 Predaceous 7 Omnivorous Saprophagous 1 Predaceous

Diplopoda

Lithobiidae 1 Geophilidae 1 Scolopendridae 1 Glomeridae 1 Craspedosomidae 1 Iulidae 1 Polydesmidae 1 Polyzoniidae 1

Isopoda 1 Araneidea 1 Acarids 1 Pseudoscorpions 1

a

Total number of cases considered with immature, post-immature or adult specimens 1 Taxa used in none analysis

379

F. Lefebvre & E. Gaudry

Table 5. Inventory of sampled Diptera. Family Calliphoridae Genus Calliphora Robineau-Desvoidy 1830 1 Chrysomya Robineau-Desvoidy 1830 1 Lucilia Robineau-Desvoidy 1830 1 Protophormia Townsend 1908 1 Phormia Robineau-Desvoidy 1830 1 Hemilucilia Brauer 1895 4 Cochliomyia Townsend 1915 1 Potamia Robineau-Desvoidy 1830 4 Hydrotaea Robineau-Desvoidy 1830 1 Ophyra Robineau-Desvoidy 1830 1 Muscina Robineau-Desvoidy 1830 1 Musca L. 1758 2 Stomoxys Geoffroy 1762 4 Pyrellia Robineau-Desvoidy 1830 4 Piophila Fallén 1810 4 Stearibia Lioy 1864 1 Fannia Robineau-Desvoidy 1830 1 Immaturea 220 89 147 41 34 1 6 45 1 34 124 35 3 1 1 1 72 59 17 16 29 1 2 4 7 1 7 14 0 7 5 3 1 18 1 2 1 3 0 0 1 1 6 2 1 1 Diet Necrophagous Necrophagous Copro-necrophagous Necrophagous Necrophagous Copro-necrophagous Necrophagous Copro-necrophagous Sapro-necrophagous Sapro-necrophagous Sapro-necrophagous Copro-necrophagous Copro-necrophagous Parasite Copro-necrophagous Sapro-necrophagous Sapro-necrophagous Sapro-necrophagous Sapro-necrophagous Sapro-coprophagous Sapro-necrophagous Phyto-Predaceous Phyto­coprophagous Sapro-coprophagous Sapro-coprophagous Sapro-xylo-zoophagous Sapro-necrophagous Sapro-phytophagous Sapro-necrophagous Sapro-necrophagous Sapro-necrophagous Sapro-coprophagous Predaceous Sapro-coprophagous Predaceous Detritivorous - Predaceous Phytophagous Sapro phytophagous Sapro-phytophagous Sapro-phyto-zoophagous Parasite Phyto-sapro-necrophagous Phyto-sapro-necrophagous Sapro-coprophagous Zoo-phytophagous Predaceous

Sarcophagidae 1 Muscidae

Piophilidae Fanniidae Phoridae 1 Sepsidae 1 Heleomyzidae 1 Cecidomyiidae 3 Micropezidae 3 Drosophilidae 4 Scatopsidae 3 Xylophagidae 4 Trichoceridae 4 Tipulidae 3 Dryomizidae 4 Sphaeroceridae 1 Syrphidae 3 Psychodidae 3 Sciomyzidae 3 Stratiomyidae 3 Scenopinidae 4 Chironomidae 3 Tephritidae 3 Tabanidae 3 Anthomyiidae 4 Ceratopogonidae 4 Tachinidae 4 Scatophagidae 4 Bibionidae 3 Anisopodidae 4 Dolichopodidae 4 Asilidae 4

a b

Total number of cases considered with immature, post-immature or adult specimens Number of cases considered with immature, post-immature specimens only 1 Taxa used both in statistical and Wagner parsimony analyses 2 Taxa used in statistical analysis only 3 Taxa used in Wagner parsimony analysis only 4 Taxa not used in analyses

380

Necrophagous insects succession

instance the analysis was not improved when it was performed using species taxonomic level data. In our study, the same statement was reliable regarding family level. Thus the taxa found on the cadavers were gathered in functional ecological groups to reduce the number of analysis units without loosing

Table 6. Inventory of sampled Coleoptera. Family Anthicidae Bostrychidae 4 Cantharidae 1 Carabidae 3 Catopidae 4 Cerambycidae 4 Cetonidae 3 Chrysomelidae 4 Cleridae 1 Coccinellidae 4 Cryptophagidae 4 Cucujidae 4 Curculionidae 4 Dascillidae 4 Dermestidae 1 Dytiscidae 4 Elateridae 1 Endomychidae 4 Geotrupidae 4 Histeridae 1 Hydrophilidae 4 Leiodidae 4 Lucanidae 4 Melandryiidae 4 Monotomidae 4 Mycetophagidae 4 Nitidulidae 1 Ostomatidae 4 Phalacridae 4 Pterostichidae 4 Ptilidae 3 Ptinidae 4 Rhipiceridae 4 Rhizophagidae 4 Scarabaeidae 1 Silphidae 1 Silvanidae 4 Staphylinidae 1 Tenebrionidae 2 Trogidae 4 Zopheridae 4

a b 4

useful data. Lastly, genus or family taxonomic levels were used as units in this study. Following the initial hypothesis, each taxon belongs to an ecological group, traditionally called a `wave'. The insects

Totala 1 1 3 17 2 1 1 1 50 1 2 1 6 1 46 2 28 1 18 70 4 2 1 1 1 1 17 1 3 1 1 2 1 1 9 83 1 114 8 3 1

Immatureb 0 0 2 5 1 0 1 0 8 0 0 0 1 1 25 0 27 0 0 14 1 0 0 1 0 0 8 0 0 0 1 0 1 0 3 40 0 26 4 0 1

Diet Saprophytophagous Xylophagous Predaceous Predaceous Saprophagous Xylophagous Xylophagous Phytophagous Predaceous Predaceous Saprophytophagous Saprophagous - Mycophagous Phytophagous Phytophagous Necrophagous Predaceous Phytophagous - Predaceous Mycophagous Coprophagous Saprophagous - Predaceous Predaceous Saprophytophagous Xylophagous Mycophagous - Xylophagous Saprophytophagous Saprophagous - Mycophagous Saprophagous Predaceous Saprophagous - Mycophagous Saprophagous - Predaceous Phytophagous Saprophagous Parasite Predaceous Coprophagous Necrophagous - Predaceous Saprophagous - Mycophagous Predaceous Saprophagous Saprophagous Saprophytophagous

Total number of cases considered with immature, post-immature or adult specimens Number of cases considered with immature, post-immature specimens only 1 Taxa used both in statistical and Wagner parsimony analyses 2 Taxa used in statistical analysis only 3 Taxa used in Wagner parsimony analysis only 4 Taxa not used in analyses

381

F. Lefebvre & E. Gaudry

Table 7. Inventory of sampled Lepidoptera. Family Geometridae 2 Noctuidae 2 Notodonidae 2 Pyralidae 1 Tineidae 1

a

Totala 1 1 1 2 8

Immatureb 0 1 1 2 7

Diet Phytophagous Phytophagous Phytophagous Saprophagous Saprophagous

Total number of considered cases with immature or post-immature or adult specimens Number of considered cases with immature or post-immature specimens only 1 Taxa used in both statistical and Wagner parsimony analyses 2 Taxa not used in analyses

b

population density on corpses was not analysed. The exact number of species found on human corpses over 12 years of activity did not provide relevant information to a forensic analysis. Methods This study was carried out in two steps. First, data were analysed using a descriptive statistic method based upon 356 cases. It could describe statistical units of species. In a second time, 86 cases with verified PMI were analysed using the Wagner parsimony method. It allowed testing the validity of the abovementioned statistical groups. Principal Component Factor Analysis (PCFA) The statistical analysis was based upon the comparison of the oldest development stages of the each specimen sampled as a function of the oldest development stage observed of a `typical group'. Three parameters within the development stage were analysed: larva, pupa and puparia. The larval stage was considered as an active phase of the immature development. The pupal or nymphal stage represented the passive phase of the immature development. The puparium or exuvia stage

Table 8. Inventory of sampled Hymenoptera. Order Hymenoptera 1 Family Braconidae 3 Chalcididae 2 Ichneumonidae 3 Plastygasteridae 2 Pteromalidae 2 Hymenoptera 1 Bethylidae 2 Formicidae 3 Myrmicidae 3

a

represented the end of the specimen developmental cycle. This method of analysis was chosen because it allowed to exclude parameters such as time and climate data. All taxa sampled were not considered. The considered taxa must belong to the cadaver fauna (necrophagous, predator, coprosaprophagous, ...). They must also be observed several times in the sampling because a single presence should avoid random sampling bias. Among the species considered, the oldest immature specimen of each species was noticed case by case (larva, pupa, nymph, puparia, exuvia). Simultaneously, the oldest developmental stage of the `typical group' was identified. Next, the development stages of the sampled species were compared taxa by taxa to the development stage identified of the `typical group'. Then, the rate of presence by species was determined. They were statistically analysed by the descriptive method of the principal component factor analysis (PCFA). The `typical group' must be defined. Sarcophagidae and Calliphoridae species (previously Calliphora sp. RobineauDesvoidy 1830, Lucilia sp. Robineau-Desvoidy 1830, Chrysomya sp. Robineau-Desvoidy 1830, Protophormia sp. Townsend 1908 and Phormia sp. Robineau-Desvoidy 1830) are commonly known to colonize a cadaver immediately after death under favourable environmental conditions (Leclercq 1978; Smith 1986; Goff 2000; Introna & Campobasso 2000; Byrd & Castner 2001; Greenberg & Kunich 2002). These species are associated with the first wave of necrophagous insects that colonized a corpse after death. They were therefore selected to define the `typical group'. The analysis was applied to 356 cases. Calculations were performed with Microsoft Excel 98® and Excel-Stat® softwares. Wagner parsimony analysis applied to synecology The aim of the Wagner parsimony analysis applied to synecology is to propose a hierarchy of a set of `objects' (`stations', `localities', `temporal successions of states that concern an object'), which is based on the distribution of `characters' - presence/absence (or abundance) distribution of organisms. This method is currently used in phylogeny through the cladistic concept (Darlu & Tassy 1993). But, unlike the cladistic, the application of the Wagner parsimony to synecology is independent of any idea of inheritance with modification. The Wagner parsimony method has been used in aquatic ecology studies (BellanSantini et al. 1994; Masselot 2002), in parasitology (Cabaret 2003) or in palaeobotany (Coiffard et al. 2007). Nel et al. (1998) theorized the application of the Wagner parsimony to synecology. We followed their concepts in the present paper. Wagner synecoparsimony sensu Nel et al. (1998) is a method of hierarchisation of `objects' based on the minimization of the total number of transformations of the character states in the hierarchy. Therefore, this method is suitable for the analysis and description of a chronological succession of necrophagous insects on a cadaver. The `characters' were the different taxa present (state `1' in the matrices) or absent (states `0') on the cadavers. Sometimes the character has three states, i.e. `0' for absence, `1' for presence of living immature stages (active and passive phases of the immature development : larva, pupa or nymph) and `2' for presence of puparium or exuvia stage representing the end of the immature developmental cycle (for precision see below). More precisely, two analyses were realized, the first one with the

Totala Immatureb 12 5 2 0 6 3 1 0 1 1 2 1 18 1 9 8 1 1 0 0

Diet Parasite Parasite Parasite Parasite Parasite Parasite Omnivorous - Predaceous Predaceous Omnivorous Omnivorous

Total number of cases considered with immature, post-immature or adult specimens b Number of cases considered with immature, post-immature specimens only 1 Taxa used in statistical analysis only 2 Taxa used in Wagner parsimony analysis only 3 Taxa not used in analyses

382

Necrophagous insects succession

`0' corresponding to the absence of collection of the concerned insect, and the second with the `0' corresponding to the real, confirmed absences and questionmarks (`?') for some couplets (object/character) in which the non collected taxon can be considered as equivocal (sensu Nel et al. 1998). In this study, it was necessary to test the hypothesis that the statistical units - defined by the PCF analysis - succeed one another over time. Nevertheless, none data correspond at `object' sensu Nel et al. (1998). The physico-chemical data of the corpse discovered could be good `objects', but they were not available in our samples. The environemental data were not filled for whole cases (tab. 2). In the database, only the PMI estimations of 86 cadavers were strictly established by feedback of police investigations. These PMI were determined as accurately as possible. Thus, they were used like `objects' sensu Nel et al. (1998) and classified by increasing PMI. Six arbitrary steps were defined (tabs. 3 and 9). Step 0 was a theoretical situation characteristic of the moment of death. No insect would be present on the corpse at step 0. This class represents outgroup that gives the primary polarization of the characters (see Nel et al. 1998). The most parsimonious tree(s) are then rooted by this outgroup. These five following steps were choosed in concordance to the current possible precisions of PMI. Morever, they allowed to define five steps with similar number of cases (about 20% of cases for each step ­ see tab. 3). This choice was done to allow the comparaison between the steps and the equal distribution of the occurrence of characters. Lastly, for each species, the oldest development stage specimen was noticed for each case. A character matrix was drawn with these collected data. The codification of the presence/absence of a specific specimen depended on each species and its development stage. Adult specimens of Diptera, Lepidoptera, and Hymenoptera were not characteristic of the diet of immature stages and were not used to determine the PMI (Amendt et al. 2007). The developmental cycle of Diptera can be described in four sequential phases: egg-laying, larval development (active phase of the immature development), pupal development (passive phase of the immature development), end of development (adult eclosion: puparium). The presence of adult specimen or the absence of immature specimen in the sample was noticed 0; the presence of at least one immature specimen (egg, larvae, pupae or nymphae) in the sample was noticed 1; the presence of at least one puparium was noticed 2. Contrary to the previous described taxa, adults and immature stages of Coleoptera can have the same biology. The presence of a single adult specimen must be taken into account because it could represent the first phase in a species colonization process before the oviposition. The development cycle can be outlined

Table 9. PMI / Step. Post Mortem Interval (PMI) Moment of death PMI < 1 week 1 week < PMI < 2 weeks 2 weeks < PMI < 1 month 1 month < PMI < 3 months PMI > 3 months Step Step 0 Step 1 Step 2 Step 3 Step 4 Step 5

in five phases: adult colonization, egg-laying: larval development (active phase of the immature development), nymphal development (passive phase of the immature development), end of development (exuvia). The absence of specimen in the sample was noticed 0; the presence of the adult specimen in the sample was noticed 1; the presence of the immature specimen (egg, larva or nymph) in the sample was noticed 2. The two Wagner parsimony exaustive analyses were performed using Paup 3.1.1.1®. The distribution of characters are analysed using MacClade 3.08a® softwares.

Results

During the period considered, 117 taxa (potential characters) were identified in the samples (tabs. 4­8 and 10).

Table 10. Presence rate of immature stages within cadaver entomofauna (%). Order Genus / Family Calliphora sp. Chrysomya sp. Cochliomyia sp. Fannia sp. Heleomyzidae sp. Hydrotaea sp. Lucilia sp. Musca sp. Muscina sp. Ophyra sp. Phoridae sp. Phormia sp. Protophormia sp. Sarcophagidae sp. Sepsidae sp. Sphaeroceridae sp. Stearibia sp. Cantharidae sp. Carabidae sp. Cleridae sp. Dermestidae sp. Elateridae sp. Histeridae sp. Nitidulidae sp. Scarabaeidae sp. Silphidae sp. Staphylinidae sp. Tenebrionidae sp. Tineidae sp. Pyralidae sp. Undetermined species La 0.86 0.94 1 0.64 0.55 0.52 0.82 1 0 0.44 0.75 0.94 0.9 0.79 0.25 1 0.31 0.5 0.8 0.17 0.1 0.19 0.31 0.12 0.33 0.26 0.3 0 0 0 1 Pb 0.07 0.06 0 0.06 0.27 0.26 0.09 0 0.33 0.3 0.25 0 0.1 0.17 0.25 0 0.23 0 0 0.17 0.14 0.33 0.31 0.25 0 0.28 0.35 0 0.14 0 0 PVc 0.07 0 0 0.3 0.18 0.22 0.09 0 0.67 0.26 0 0.06 0 0.04 0.5 0 0.46 0.5 0.2 0.66 0.76 0.48 0.38 0.63 0.67 0.46 0.35 1 0.86 1 0

Diptera

Coleoptera

Lepidoptera Hymenoptera

a b

L = Larva = active immature stage P = Pupa = passive immature stage c PV = Puparium

383

F. Lefebvre & E. Gaudry

1. Statistical analysis Tab. 10 showed the inventory of the presence rate of the oldest immatures stages case by case and taxa by taxa. They were determinated by comparaison with the development stage of the `typical group' according to three parameters larval (L), pupal (P) and puparium or post-pupal (PP) (cf. § Material & Methods). Dipterans. Calliphoridae larvae were usually observed when the oldest immature stage of the `typical group' is larval. For example, when immature specimens of Calliphora sp. were observed on a cadaver, the oldest development stage identified of the `typical group' was larval (86%), pupal (7%) and post-pupal (7%). The same trend was observed for Chrysomya sp. (94%, 6%, 0%), Lucilia sp. (82%, 9%, 9%), Protophormia sp. (90%, 10%, 0%), and Phormia sp. (94%, 0%, 6%). The presence rate of Sarcophagidae immature stage observed simultaneously with the larval stage of `typical group' was 79%, 17% with the pupal stage and 4% when the development of typical group was reached. The results observed between the different developmental stage for these species and this theorical `typical group' seemed coherent. Similar distribution was found with other taxa. Immature specimens of Phoridae were mainly sampled along with the larval stage (75%) and pupal stage (25%) of the `typical group'. Coleopterans. Regarding Coleoptera, the data showed that Tenebrionidae and Dermestidae families could be associated with the end of development cycle of the first generation of the `typical group' (puparium), with respectively 100% and 76%. The species of Carabidae were mainly associated with the larval stage (80%), but never with the pupal stage and in 20 % of cases with the achievement of the pupal stage of the `typical group'. However, the affinities of the `typical development stage' with a great number of species cannot be easily interpreted. These latters could be associated with the three states of development of the `typical group'. For example, Ophyra sp. Robineau-Desvoidy 1830 and

Table 11. Statistical units described by PCFA. Statistical Units Unit 1

Stearibia sp. Lioy 1864, were respectively sampled in 44% and 31% simultaneously with the larval `typical group, 30% and 23% with pupal `typical group', 26% and 46% with the post-pupal `typical group'. This observation was similar for immature specimens of Staphylinidae and Histeridae (Coleoptera) respectively associated in 30% and 31% with larval `typical stage', 35% and 31% with pupal `typical group', 35% and 38% with post-pupal `typical group'. The single observation of the presence rate of species in comparison to `typical group' can not explain the distribution of the data. Data were then studied by a PCFA. Results were shown in fig. 1 and tab.11. This diagram showed that two principal components were sufficient to describe 100% of data variability. These two main axes could explain the whole data distribution. F1 axis could represent the presence or absence of the immature specimens of the `typical group' on the corpse. It explained almost 70% of data variability. F2 axis could represent the active and passive development phases of these specimens. It explained the remaining 30% of data variability. Finally, three statistical units could be highlighted (tab. 11). These statistical units 1-3 were defined according to the vectors representing the different development stages of the typical group. The first unit represented the taxa statically sampled when the typical group was present on the corpse at larval stage (L). This statistical unit 1 was logically compound of Diptera : Sarcophagidae and Calliphoridae (Lucilia sp., Calliphora sp., Chrysomya sp., Protophormia sp. and Phormia sp.). Within this order, Muscidae (Musca sp. L. 1758), Fanniidae (Fannia sp. Robineau-Desvoidy 1830), and Sphaeroceridae were identified as some Coleopterans and Hymenopterans species. The presence of Musca sp. (Muscidae) was coherent with the biology of the species (housefly). In natural conditions, eggs were laid mainly on excrement or decaying vegetables. However, the oviposition could occur on fresh human corpse with or without excrement contamination (Smith 1986). Sometimes,

Unit 2 Unit 3

Taxa Diptera: Sarcophagidae, Calliphoridae (Calliphora sp., Lucilia sp., Chrysomya sp., Protophormia sp., Phormia sp., Cochliomya sp.), Fanniidae (Fannia sp.), Muscidae (Musca sp.), Sphaeroceridae Coleoptera: Carabidae Hymenoptera Diptera: Muscidae (Ophyra sp., Hydrotaea sp.), Piophilidae (Stearibia sp.), Sepsidae, Heleomyzidae Coleoptera: Staphylinidae, Histeridae, Silphidae, Elateridae Coleoptera: Cleridae, Dermestidae Lepidoptera: Tineidae

384

Necrophagous insects succession

the Musca species were included in the first wave of Mégnin's classification (Leclercq 1978). Fannia sp. was also observed in this statistical unit. Although these species were considered in the fourth wave (Introna & Campobasso 2000; Byrd & Castner 2001; Greenberg & Kunich 2002; Wyss & Cherix 2006), they were also observed on a carrion about 4­10 days after death (Payne 1965; Wasti 1972). Sphaeroceridae, usually associated with the fourth wave, were recorded in the first statistical group. They are reputed to be indoor species known to colonize dung and excrement. They could colonize carrion but are never related to a time parameter (Smith, 1986). Their presence in this group could be explained by the discovery of cadavers under particular environmental conditions (wrapped, ...). Considering Coleoptera, Carabidae was the single taxa recorded in the statistical unit 1. The species belonging to this family could be found on carrion as predators that could explain their presence. Their forensic interest is then limited (Smith 1986). All Hymenopterans species were gathered in a single group to increase the occurrence number. They usually are parasites of other insects. Considering that hosts are mainly necrophagous larvae, they can be indirectly related to the presence of a decaying corpse. Their development depends on the larval stage of the `typical group'. The statistical unit 2 collected taxa sampled on a cadaver simultaneously with the pupal stage of the `typical group' (P). In plain language, these taxa were statistically present on a corpse when specimens of the `typical group' had reached the end of active immature development stage (statistical unit 1). The statistical unit 2 showed Diptera and Coleoptera species. Considering Diptera, Ophyra sp. RobineauDesvoidy 1830, Hydrotaea sp. Robineau-Desvoidy 1830 and Stearibia sp. Lioy 1864 were traditionally associated with the fourth or fifth wave (Leclercq 1978; Smith 1986; Byrd & Castner 2001). Larvae of Sepsidae and Heleomyzidae were found on excrement and decomposing organic matter (vegetal or animal). These species could be found on the carrion during caseic or amnoniacal fermentation along with Piophilidae (Leclercq 1978; Smith 1986). Staphylinidae, Histeridae, Silphidae and Elateridae (Coleoptera) were present in the statistical unit 2. Excepted Elateridae, species of these families are listed in the fifth wave (Mégnin 1894; Leclercq 1978; Smith 1986; Wyss & Cherix 2006). The statistical unit 3 collected taxa that were mainly observed when the oldest specimen of the `typical group' had finished their development cycle (puparium stage,

PP). Results showed that it was a heterogeneous unit composed by Coleoptera and Lepidoptera species. Immature Cleridae and Dermestidae (Coleoptera) were observed in this unit. Dermestidae larvae are necrophagous and feed on a wide variety of dried animal matters. These species are listed in third wave, when the corpse is completely dry. Cleridae are predaceous (entomophagous) either as adult or larva and are usually associated with immature Dermestidae (Lepesme 1944). Tineidae (Lepidoptera) were also found in this statistical unit 3. This taxon is logically associated with Dermestidae (Smith 1986) and listed in the third wave. Its occurrence in the third statistical unit is coherent. 2. Ungrouped taxa In our results, some taxa could not be classified according to the development stage of the `typical group'. Dipterans. For example, Muscina sp. RobineauDesvoidy 1830 was traditionnally listed in the first wave (Leclercq 1978). However, the statistical analysis did not confirm this statement. Indeed, immature specimens of Muscina were not clearly associated with the larval stage of the `typical group'. This equivocal position of these taxa could be explained by their biology. These species are usually associated with fresh corpses, but Muscina larvae seem to have a preference for human faeces and are not commonly found on carrion (Smith 1986). A similar phenomon was observed with Phoridae, usually associated with the fifth wave. Within this family, Conicera sp. Meigen 1830, Megaselia sp. Rondani 1856, are found on carrion. Nevertheless, the relation between the presence of Phoridae species and a necrophagous species community (Smith 1986) is not obvious. In addition, Phoridae (so called coffin flies) are associated with the necrophagous fauna found on the buried carrion (Introna & Campobasso 2000; Byrd & Castner 2001; Greenberg & Kunich 2002). The small amount of cases of buried cadavers in the database may have impacted the statistical analysis (tab. 2). Coleopterans. Nitidulidae, Cantharidae, Scarabaeidae and Tenebrionidae were not associated with any statistical group. Cantharidae are not described as taxa of forensic importance (Smith 1986; Byrd & Castner 2001). Their presence on carrion was probably due to the opportunist behaviour of the great majority of beetles. Nevertheless, Nitidulidae, Scarabaeidae, and Tenebrionidae belong to the cadaver fauna. Tenebrionidae, are usually found in the eighth wave, on completely dried corpses (Mégnin 1894; Smith 1986). Nitidulidae and Scarabaeidae are known to colonize

385

F. Lefebvre & E. Gaudry

Table 12. Data matrix. 1 Calliphoridae 2 Sarcophagidae 3 4 Piophilidae 5 6 7 Heleomyzidae 8 Micropezidae 9 10 11 Stratiomyidae 12 Chironomidae 13 Staphylinidae 14 15 16 17 Nitidulidae 0 0 0 0 1 1 18 Cantharidae 0 0 0 0 1 1

Histeridae 0 0 1 1 1 0

Syrphidae

Period 0 Period 1 Period 2 Period 3 Period 4 Period 5

0 1 1 2 2 2

0 1 1 2 0 2

0 0 1 1 2 2

0 0 1 0 2 2

0 0 1 1 2 2

0 0 0 0 2 2

0 0 0 0 1 2

0 0 0 0 1 1

0 0 1 0 1 1

0 0 0 1 1 1

0 0 1 0 0 1

0 0 1 0 1 0

0 0 1 1 1 1

0 0 1 1 2 1

1: presence ­ 0: absence

carrion at an advanced stage of decomposition too. However, they could not be associated with a specific wave. Lepidopterans. The position of Pyralidae could be explained by the weak occurrence of this taxon in the samples. But we could not estimate a rate of

appearance on a human corpse (tab. 6). Moreover, the forensic importance of this taxon has not been clearly highlighted. Indead, some authors associated it either with the third and seventh waves (Mégnin 1894; Leclercq 1978), or with third wave only (Smith 1986; Introna & Campobasso 2000). Others did not

Figure 1 Distribution of taxa in function of typical development stages.

386

Elateridae 0 0 1 1 1 1

Fanniidae

Tipulidae

Muscidae

Silphidae

Phoridae

Necrophagous insects succession

consider the Lepidoptera group as a taxon of forensic importance (Byrd & Castner 2001; Wyss & Cherix 2006). 3. Wagner parsimony analysis The previous statistical analysis allowed describing our data in three homogeneous units. These three statistical groups could then be compared with three development stages of a theoretical ecological group. Characteristic pairs of taxa present or absent among 86 concrete cases were examined by the Wagner parsimony method, to confirm these observations. This analysis could test whether a modification in the insect population could characteristize changes or steps in the decomposition of a corpse over time. Presences/ absences of species were studied at six periods of time from the recent death up to more than three months (tab. 9). Taxa/characters had to be found at least at two periods of time, otherwise they can be considered as non informative characters (autapocoenose sensu Nel et al. 1998). In order to obtain a homogeneous pattern, taxa were studied at the family taxonomic level: Calliphoridae (Calliphora sp., Lucilia sp., Phormia sp., Protophormia sp. and Chrysomya sp.), Muscidae

(Ophyra sp., Hydrotaea sp.), Piophilidae (Stearibia sp.), and Fanniidae (Fannia sp.). 18 taxa (i.e. characters) were compiled and analysed (tab. 12). The program Paup is based on the Wagner parsimony method (cf. § Material & Methods) and performed the data analysis. The first analysis (exhaustive research), using `0' for all absence of collects, gave a single most parsimonious tree (fig. 2). The matrix of the corresponding periods and taxa/ characters is given in tab. 12. This hierarchy is 30 steps long, has a consistency index (CI = 0.83) and a high retention index (RI = 0.78). In this analysis, the absence of immature individuals of Tipulidae, Chironomidae, and Piophilidae at the period 3, Stratiomyidae (periods 3 & 4) and Sarcophagidae (period 4), introduced homoplasies in the analysis. Some of these homoplasies correspond to the presence of immature individuals of Stratiomyidae and Tipulidae (Diptera) in the samples. Sarcophagidae and Piophilidae are well known to be present on decaying corpses (Smith 1986). The situation is similar with chironomid species, whose aquatic larvae can live on immersed corpses. Some authors reported that Stratiomyidae, like Hermetia illucens (L. 1758), may

Figure 2 Most parcimonious tree.

387

F. Lefebvre & E. Gaudry

have a forensic importance (Tomberlin et al. 2005). In order to optimize the analysis, we tested the hypothesis that `0' for these taxa correspond to random lack of sampling. As we can not exclude possible presence we codified these data with question marks `?'. The second analysis (exhaustive research) was performed using some question marks for ambiguous absences of collection (tab. 13). The optimisations of the missing character states (`?') were obtained using the accelerated transformation method (ACCTRAN). `?' concerning Sarcophagidae, Piophilidae, Tipulidae, Stratiomyidae and Chironomidae (respectively characters 2, 4, 9, 11, 12) can be optimized as potential presences. The program Paup gave the same most parcimonious tree with a different distribution of the character states (Fig. 2). This hierarchy is 28 steps long, has a high consistency index (CI = 0.89) and a high retention index (RI = 0.82). The first monocoenotic group (sensu Nel et al., 1998) of periods (1+2+3+4+5) was strongly supported by the presence of immature specimens of Sarcophagidae and Calliphoridae (Calliphora sp., Chrysomya sp., Lucilia sp., Protophormia sp. and Phormia sp.) on corpses (characters 1 and 2). These taxa, colonizing a corpse immediately after death, are traditionally classified in the first wave. The second monocoenotic group of periods (2+3+4+5) corresponded to a PMI over one week. This stage is supported by the presence of immature specimens of Diptera: Muscidae (Ophyra sp., character 3), Fanniidae (Fannia sp., character 5) and Coleoptera: Staphylinidae (character 13), Silphidae (character 14), Histeridae (character 15) and Elateridae (character 16). The third monocoenotic group of periods (3+4+5) was mainly characterised by the end of development

Table 13. Data matrix optimised. 1 Calliphoridae 2 Sarcophagidae 3 4 Piophilidae 5 6 7 Heleomyzidae 8 Micropezidae

cycle of the first necrophagous insects (presence of puparia of the previous species). These taxa belonged to the Calliphoridae (character 1) and the Sarcophagidae (character 2). Some immature specimens of Syrphidae (Diptera, character 10) were also present. The fourth monocoenotic group (4+5) was defined by the end of the development cycle of the muscid flies (character 3), Piophilidae (character 4), and Fanniidae (character 5). This group could equally be characterised by the presence of immature specimens of Heleomyzidae (character 7), Micropezidae (character 8), and the nitidulid beetles (character 17), Cantharidae (character 18). Lastly, periods 4 and 5 were both characterised by an autapocoenose, in other words, the last taxa colonizing a corpse. Period 4 corresponded to the end of the life cycle of Silphidae (character 14), whereas period 5 correponded to the end of the Heleomyzidae cycle (character 7). These were interesting taxa supporting the association at different periods. When this above-mentioned list is compared with the biology of species, it appears that the necrophagous insect population could be characterised by the development stages of several taxa: Calliphoridae (Calliphora sp., Chrysomya sp., Lucilia sp., Protophormia sp., Phormia sp.), Sarcophagidae, Muscidae (Ophyra sp., Hydrotaea sp.), Piophilidae (Stearibia sp.), Fanniidae (Fannia sp.), Staphylinidae, Histeridae, and Silphidae. Indeed, the observation of the associations of their different development stages allows defining four different `necrophagous groups' within insects population. The Wagner analysis showed that the composition of the insect population on corpse at a moment labelled "t" could be characterised by the presence of specific taxa at defined development stages. We proposed to

9

10

11 Stratiomyidae

12 Chironomidae

13 Staphylinidae

14

15

16

17 Nitidulidae 0 0 0 0 1 1

18 Cantharidae 0 0 0 0 1 1

Histeridae 0 0 1 1 1 0

Syrphidae

Period 0 Period 1 Period 2 Period 3 Period 4 Period 5

0 1 1 2 2 2

0 1 1 2 ? 2

0 0 1 1 2 2

0 0 1 ? 2 2

0 0 1 1 2 2

0 0 0 0 2 2

0 0 0 0 1 2

0 0 0 0 1 1

0 0 1 ? 1 1

0 0 0 1 1 1

0 0 1 ? ? 1

0 0 1 ? 1 0

0 0 1 1 1 1

0 0 1 1 2 1

1: presence ­ 0: absence - ?: missing data

388

Elateridae 0 0 1 1 1 1

Fanniidae

Tipulidae

Muscidae

Silphidae

Phoridae

Necrophagous insects succession

define four `necrophagous groups' as four sets of taxa that support the four monocoenotic groups of periods mentioned above. Each group characterises one monocoenotic group of periods and should correspond to a step in the succession pattern of insects population on a corpse. The first `necrophagous group' is characterised by the oviposition of the pioneers on a cadaver. Immature specimens of the species of the first wave (Calliphoridae and Sarcophagidae) were found. The second `necrophagous group' is characterised by addition of immature stages of Muscidae, Fanniidae, Piophilidae, Staphylinidae, or Silphidae. The third `necrophagous group' is observed when some of the oldest specimens of the first `necrophagous group' have ended their development cycle. Simultaneously, the development of the second `necrophagous group' goes on. The fourth `necrophagous group' is characterised by the end of the development cycle of the oldest specimen of Ophyra sp. (Muscidae) and Fannia sp. (Fanniidae) In concrete terms, a single specimen of each `necrophagous group' can characterise a population of its period. Thus, if only larvae of Lucilia sp. and Calliphora sp. are found on a corpse, then the population is classified at period 1. If immature stages of Calliphora sp., Lucilia sp. and Ophyra sp. are simultaneously sampled, then the population is classified at period 2. If larvae of Ophyra sp. are present at same time as Lucilia sp. puparia, then the population is classified at period 3. Lastly, if only Lucilia sp., Calliphora sp. and Ophyra sp. puparia are present, then the population is classified at period 4. This reasoning requires caution because the oldest specimens of each species are only considered. Within a same species, some individuals have finished their development cycle whereas development of younger larvae is still ongoing. In respect with the forensic entomological analysis, the developments of these studied species are considered to be ended (Amendt et al. 2007). If direct physico-chemical parameters of the states of decay of cadavers were available, the hypothesis of a

Table 14. Taxa supporting PMI groups. Group Period 1+2+3+4+5 Period 2+3+4+5 Period 3+4+5 Period 4+5 Calliphoridae L, Sarcophagidae L

succession pattern of the insect populations on a corpse could be tested. They could be compared as attributes with the most parsimonious tree (Nel et al. 1998). It would allow comparing concrete medical informations with different steps of insect populations on corpses. Unfortunately, such parameters are not available in our database.

Discussion

The statistical and Wagner parsimony analyses showed complementary results. The statistical analysis of the presence rate of the taxa on corpses, according to the oldest typical developmental stage, showed that three statistical units could be defined (tab. 11). Species of the first unit are mainly Diptera species corresponding to the first `wave' sensu Mégnin 1894; Leclercq 1978; and Smith 1986. Species of the second unit are statistically associated with the pupal stage of the typical group. Immature specimens of the third unit species are usually found on corpses when the oldest Diptera of the first group were represented by puparia states. The Wagner parsimony analysis allowed defining four different steps in the cadaver population pattern. The latter were characterised by a succession of several species at different development stages over time (tab. 14). Overall, the comparison of statistical and Wagner parsimony results supported the hypothesis of a modification of the necrophagous insects population present throurought the decaying process. The colonization of a corpse begins with the egg-laying (or larviposition) of species belonging to the first group. This is the first step of this dynamic of population. The step 1 of the Wagner analysis can be characterised by the presence on the cadaver of the statistical unit 1, without Musca sp. (Muscidae) not studied in the Wagner parsimony analysis, immature stages of Calliphoridae: Calliphora sp., Lucilia sp., Chrysomya sp., Phormia sp., Protophormia sp. and Sarcophagidae (figs. 1­2, tabs. 11 and 14). The step 2, reached when the species of the second

Taxa - Characters Muscidae L, Piophilidae sp. L, Fanniidae L, Chironomidae L, Staphylinidae L, Silphidae L, Histeridae L, Elateridae L Calliphoridae Pv, Sarcophagidae Pv, Syrphidae L Muscidae Pv, Piophilidae Pv, Fanniidae Pv, Phoridae Pv, Heleomyzidae L, Micropezidae L, Nitidulidae L, Cantharidae L

L = Larva - P = Pupa - Pv = Puparium

389

F. Lefebvre & E. Gaudry

group colonize the cadaver, is mainly characterised by arrival of Ophyra sp., Stearibia sp., Fannia sp., Staphylinidae, Silphidae, and Histeridae larvae. Excepted Fannia sp., this group matches with unit 2 in the statistical analysis (figs. 1­2, tabs. 11 and

14). Usually, immatures stages belonging to the unit 2 are associated with the end of active immature development stage of the unit 1. But, Fannia sp. seems to be statistically associated with the immature stages of the described pioneer species. Thus, it is possible that

Figure 3 Dynamic of necrophagous insect population

390

Necrophagous insects succession

the Fannia specimens occur on the cadaver before most of the species of this ecological group (second group). The step 3 of the necrophagous insect population occurs at the end of the life cycle of some of the oldest specimens of the first species group. The step 4 and 5 are mainly characterised by the end of development of most Diptera species of the second group. The third statistical unit is characterized by the association of Dermestidae, Cleridae, and Tineidae larvae with the end of the development of the typical group (fig. 1, tab. 11). But, this unit 3 cannot directly be associated with step 3, 4 or 5. When a corpse is discovered, forensic entomology can help to estimate a minimum or a maximum PMI (Amendt et al. 2007). The minimum PMI is determinated using the pioneer species, the maximum PMI must be deduced from a chronological succession pattern. The present results could help to the estimation of this maximum PMI following the hypothesis of a dynamic of the necrophagous insect populations over time. It allows determining an upper and a lower limit of the insect succession pattern on the corpse (fig. 3). The lower limit assumes that the first egg-layings of the second group are considered at the end of active immature development stage of the oldest specimens of the first group. This `minimal hypothesis' could be suitable when a corpse is discovered under optimal environmental conditions for insect activity. The upper limit supposes that the first egg-laying of the second group occurs during the emergence of the oldest specimens of the first group. This reasoning could be applied when environmental conditions are unfavourable to insect activity. Thus, the lower limit allows determining a shorter maximum PMI. The time of developmement of the single first group of necrophagous insects on the corpse is reduced. The second group could colonize the cadaver at the end of active immature development stage of the oldest specimens from the first group has started: we called it `short succession'. On the other hand, the upper limit allows determining a longer maximum PMI. The second group could colonize the corpse at the end of the pupal stage (adult emergence) of the oldest specimens from the first group: we called it `long succession' as the time of presence of the single first group of necrophagous insects is extended

populations on human cadavers over time, based upon analyses of data from real cases. The succession phases within insect population are characterized by species associations and the development stages. Three different necrophagous groups and four steps in the dynamic of their population were defined by the present study. PMI estimations seemed to be theoritically reliable whatever the specific environmental conditions, the location and period of discovery of cadavers. However, this succession pattern of taxa on the corpse must be considered as a preliminary result. Each new case analysed by the forensic entomology department is susceptible to provide additional information to update the database by inclusion of every species identified in the matrix. By continuous updating, the database would be an interesting tool to test the present hypothesis of succession and make it more and more reliable. Thanks to this analytical method, the increase of the matrix is unlimited.

Acknowledgments. The authors thank J. Hebrard, former Director of the Forensic Science Institute of the French Gendarmerie (IRCGN). They are very grateful to the former and actual fellow workers of the Forensic Entomology Department (IRCGN): Ph. Masselin, J. Salon , B. Ceccaldi, C. Rocheteau, J.-M. Vian, Y. Malgorn, J.-B. Myskowiak, B. Chauvet, T. Pasquerault, B. Vincent and L. Dourel. Special thanks to A. Nel, G. Masselot for critical reviews of the manuscript and S. Staunton, F. Mille, for their precious help.

References

Amendt J., Campobasso C. P., Gaudry E., Reiter C., LeBlanc H. N., Hall M. J. R. 2007. Best practice in forensic entomology - standards and guidelines. International Journal of Legal Medicine 121 (4): 90-104. Anderson G. 2001. Insect Succession on Carrion and Its Relationship to Determining Time of Death, p. 143-175. in: Byrd J. H., Castner J. L. (eds), Forensic Entomology: The Utility of Arthropods in Legal Investigations. CRC Press, Boca Raton, FL, USA. Bellan-Santini D., Dauvin J. C., Bellan G. 1994. Analyse de données en écologie benthique : utilisation de la méthode de l'analyse de parcimonie. Oceanologica Acta 17 (3): 331-340. Byrd J. H., Castner J. L. 2001. Insects of Forensic Importance, p. 43-79 in: Byrd J. H., Castner J. L. (eds), Forensic Entomology: The Utility of Arthropods in Legal Investigations. CRC Press,. Boca Raton, FL, USA. Cabaret J. 2003. Relating Parasite Communities to Host Environmental Conditions using Phylogenetic Tools. Parasite 10: 287-295. Coiffard C., Gomez B., Thevenard F. 2007. Early Cretaceous Angiosperm Invasion of Western Europe and Major Environmental Changes. Annals of Botany 100: 545-553. Darlu P., Tassy P. 1993. Reconstruction Phylogénétique : Concept et Méthodes. Collection Biologie Théorique, Masson, Paris, France, 245 p. Goff M. L. 2000. A Fly for the Prosecution : How Insect Evidence Helps Solve Crimes. Harvard University Press, London, 225 p. Greenberg B., Kunich J. C. 2002. Entomology and the Law: Flies as Forensic Indicators. University Press, Cambridge, 306 pp. Introna F., Campobasso C. P. 2000. Forensic Entomology, p. 793-846 in: Papp L., Darvas B. (eds.), Contributions to a Manual of Palaearctic Diptera (Vol. 1). Science Herald, Budapest, Hungary.

Conclusion

This study proposed a new approach and a hypothesis on the dynamic of the necrophagous insects'

391

F. Lefebvre & E. Gaudry

Johnston W., Villeneuve G. 1897. On the Medico-Legal Application of Entomology. Montreal Medical Journal 26: 81-90. Leclercq M. 1978. Entomologie et Médecine Légale : Datation de la mort. Collection de Médecine Légale et de Toxicologie Médicale. Masson, Paris, 100 p. Lepesme P. 1944. Les Coléoptères des denrées alimentaires et des produits industriels entreposés. Paul Lechevalier Editeur, Paris, 335 p. Masselot G. 2002. La synécoparcimonie : un outil d'évaluation biologique de la qualité des eaux courantes : Théorie et applications. Sciences de la Vie et de la Nature, Museum National d'Histoire Naturelle, Paris, 451 p. Mégnin J. P. 1894. La Faune des Cadavres : Application de l'entomologie à la Médecine Légale. Encyclopédie Scientifique des Aide-Mémoire, Masson, Gauthiers-Villar et Fils, Paris, 214 p. Motter M. G. 1898. A Contribution to the Study of the Fauna of the Grave. A Study of One-Hundred and Fifty Disinterments with Some Additional Experimental Observations. Journal of the New-York Entomological Society 6: 201-231. Nel A., Nel J., Masselot G., Thomas A. 1998. An investigation into the application of the Wagner parsimony method in synecology. Biological Journal of the Linnean Society 65: 165-189.

Payne J. A. 1965. A summer carrion study of the baby pig Sus crofa Linnaeus. Ecology 46: 592-602. Schloenly K., Goff M. L., Early M. 1992. A BASIC Algorithm for Calculating the Postmortem Interval from Arthropod Successional Data. Journal of Forensic Sciences 37 (3): 808-823. Smith K. G. V. 1986. A Manual of Forensic Entomology. British Museum (Natural History), London, United Kingdom, 205 p. Tomberlin J. K., Sheppard D. C., Joyce J. A. 2005. Black soldier fly (Diptera: Stratiomyidae) colonization of pig carrion in South Georgia. Journal of Forensic Sciences 50 (1): 152-153. Wasti W. S. 1972. A study of the carrion of the common fowl, Gallus domesticus, in relation to arthropod succession. Journal of the Georgia Entomological Society 7: 221-229. Wyss C., Cherix D. 2006. Traité d`Entomologie Forensique : Les Insectes sur la scène de crime. Presses polytechniques et universitaires romandes, Lausanne, 317 p. Yovanovitch G. P. 1888. Entomologie appliquée à la Médecine Légale. Librairie Ollier-Henry, Paris, 132 p.

392

Information

ASEF 3-2009.indb

16 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

409325


Notice: fwrite(): send of 209 bytes failed with errno=104 Connection reset by peer in /home/readbag.com/web/sphinxapi.php on line 531