Appendix 6 – Insect remains Eileen Reilly Botany Department, Trinity College Dublin, College Green, Dublin 2 137 APPENDIX 6: CONTENTS LIST OF ILLUSTRATIONS . 140 LIST OF TABLES . 141 1 INTRODUCTION . 142 2 METHODOLOGY . 143 3 ANALYSIS. 147 4 DISCUSSION . 149 139 LIST OF ILLUstrations 6.1 Graph of habitat groups indicated for all insects identified – raw counts and percentage presence of each habitat group (source: created by BUGSCEP) . 144 6.2 Summary data – total number of species, individuals and sum abundance per sample for insect remains identified (source: created by BugsCEP). 146 6.3 NMS ordination showing 1-axis solution for samples examined from insert remains . 149 6.4 NMS ordination showing 1-axis solution for species distribution. 150 140 LIST OF TABLES 6.1 Species list for insects (source: nomenclature follows Lucht 1987) . 152 141 1 INTRODUCTION Six samples were selected for insect analysis. Insect Samples for insect analysis were selected from analysis is just one strand of a multidisciplinary the pre-ditch ground surface, possible old stream approach being taken to understand the local envir- channel and fills from two phases of the medieval onment of the area, which also includes pollen, bone, ditch (see Section 3 below for full description). mollusca, wood and plant-macrofossil analyses. 142 2 METHODOLOGY Samples were 10 litres in volume. Sub-samples of time, in the case of stratified sequences, or spatial 4–5 litres were analysed. The remaining material analysis in the case of dispersed archaeological was retained for potential future use. samples. The package uses 22 EcoCodes derived All samples were subject to the paraffin flotation from the work of various authors (Kenward 2001; method of extracting insect remains from archaeo- Ponel 1995; Robinson 2001) and the ecological codes logical deposits, devised first by Coope and Osborne assigned by Koch (1989–92). Each taxon (or individ- (1968) and later modified by Kenward (1980), ual) may be assigned to more than one habitat but with further refinements by Kenward et al (1986). will only be assigned to one ‘indicator’ class, where The samples were disaggregated using water and appropriate. Illus 6.1 shows two EcoGraphs, one dis- washed over a 300μm sieve. The retent in the sieve playing the raw counts per habitat group and the was then treated with paraffin and cold water added. other the percentage presence of each habitat as a The paraffin concentrates the insects by adhering to proportion of the total number of taxa (‘environmen- the waxy cuticle of the insect exoskeleton. The flot tal representations’) in that sample. Illus 6.2 shows was then poured through a 300μm sieve, washed the summary statistics for the data. The analysis with detergent in hot water to remove the excess and discussion (Sections 3 and 4) are based on the paraffin and stored in 70% ethanol. Processing took results of these graphs. place at the flotation facility of Margaret Gowen and The index of diversity for the assemblages was Company Ltd, Merrion Square, Dublin 2. also calculated (see Section 4 below). Fisher’s alpha Flots varied in size between 100 and 200ml. All is a mathematical model used to measure diversity were fully sorted and insect sclerites extracted in ecological communities (Fisher et al 1943). onto wet filter paper. Identifications were carried In addition, NMS (non-metric multi-dimensional out using published keys and the writer’s own scaling) ordination of the data was carried out, using reference material. Table 6.1 details the full species PC-Ord 5.0 (McCune & Grace 2002). Ordination is list. Species marked with ‘?’ indicate identifications used to summarise complex relationships, extract- that require further checking. The species list was ing one or a number of dominant patterns from an then entered into Bugstats, part of BugsCEP, the infinite number of possible patterns ibid( ). It is used Coleopteran Ecology Package devised by Philip and here to examine underlying variance/similarities in Paul Buckland (Buckland & Buckland 2006). This the insect assemblages between deposit types (illus allowed for analysis of the insect data by producing 6.3 & 6.4). NMS is well suited to data that are non- a multi-sample environmental summary diagram normal or on arbitrary, discontinuous, or otherwise using coded habitat data. This diagram should aid questionable scales, particularly data sets that have identification of environmental changes through upwards of 70% zeroes. 143 Sample Standing water Open wet Wetlands/ Pasture/ Heathland & Wood and trees Dry dead wood Disturbed/ Sandy/dry Dung/foul Mould beetles Dung Numbers Aquatics habitats marshes Halotolerant Dung Meadowland moorland arable disturbed/ habitats Carrion Synanthropic arable S67 S86 S68 raw counts per habitat group S56/55 S110 05101520 0 5101520 0 5101520 0 5 101520 -1 0 0 5 101520 0 5 101520 0 5 101520 0 5 101520 0 5 10 15 20 0 5 101520 0 10 20 0 5 101520 0 5 101520 0 5 101520 0 5 10 15 20 0 5 10 15 20 0 5 10 15 20 Standing water Open wet Wetlands/ Halotolerant Dung Pasture/ Heathland & Wood & trees Dry dead wood Disturbed/ Sandy/dry Dung/foul Mould beetles Synanthropic Samples Aquatics Meadowland arable disturbed/ Carrion habitats marshes Dung moorland habitats arable S67 S86 S68 % representation per habitat group S56/55 S110 0 5 1015 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 0 5 1015 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 -1 0 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 Illus 6.1 Graph of habitat groups indicated for all insects identified – raw counts and percentage presence of each habitat group (source: created by BUGSCEP) 144 Sample Standing water Open wet Wetlands/ Pasture/ Heathland & Wood and trees Dry dead wood Disturbed/ Sandy/dry Dung/foul Mould beetles Dung Numbers Aquatics habitats marshes Halotolerant Dung Meadowland moorland arable disturbed/ habitats Carrion Synanthropic arable S67 S86 S68 raw counts per habitat group S56/55 S110 05101520 0 5101520 0 5101520 0 5 101520 -1 0 0 5 101520 0 5 101520 0 5 101520 0 5 101520 0 5 10 15 20 0 5 101520 0 10 20 0 5 101520 0 5 101520 0 5 101520 0 5 10 15 20 0 5 10 15 20 0 5 10 15 20 Standing water Open wet Wetlands/ Halotolerant Dung Pasture/ Heathland & Wood & trees Dry dead wood Disturbed/ Sandy/dry Dung/foul Mould beetles Synanthropic Samples Aquatics Meadowland arable disturbed/ Carrion habitats marshes Dung moorland habitats arable S67 S86 S68 % representation per habitat group S56/55 S110 0 5 1015 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 0 5 1015 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 -1 0 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 Illus 6.1 Graph of habitat groups indicated for all insects identified – raw counts and percentage presence of each habitat group (source: created by BUGSCEP) 145 Sum representation of individuals Sample Abundance (raw counts) Number of species in all habitat categories Numbers S67 S86 S68 S56/55 S110 -1 0 0 25 50 75 100 125 150 0 25 50 75 0 25 50 75 100 125 150 175 Illus 6.2 Summary data – total number of species, individuals and sum abundance per sample for insect remains identified (source: created by BugsCEP) 146 3 ANALYSIS Samples are analysed in chronological order. (Limnebius sp., Helophorus sp.) and one possible indicator of flowing water, Oulimnius sp., present. The latter beetle is generally found under stones or 3.1 Context [142], sample 110 on moss in fast-flowing, shallow streambeds (Koch 1989–92) and is one of the few species present that Context 142 is described as a blackish-brown peaty suggests the presence of a stream in the locality. A deposit, gritty and organic. The clay and charcoal number of species present suggest that, once again, content was high. Animal bone and shellfish were dung/foul-indicating habitats prevailed. The carrion visible during processing. feeder Necrophorus humator was identified, as well This produced an interesting, albeit small, assem- as Aphodius spp., Anotylus tetracarinatus and A. blage (MNI = 41, illus 6.1 & 6.2, Table 6.1). Many complanatus. These beetles probably represent the of the sclerites were fragmented and identifica- beginnings of accumulated rubbish associated with tion past genus level was not always possible. This the growth of the medieval town. In addition, the resulted in a mixed signature within the assemblage, disturbed waste ground and/or arable signature with many species assigned to multiple habitats. present in context [142] is also visible in this assem- However, generally, foul habitats/dung/pasture blage through species such as Otiorhynchus sulcatus, indicators were very common, as well as wetland/ Chrysolina sp. and Phyllotreta sp. (Table 6.1). O. marsh indicators. The most commonly encountered sulcatus (the ‘vine’ weevil) is generally closely asso- beetles were Helophorus spp., generalist water ciated with humans (Morris 1997). Along with other beetles found in a wide variety of aquatic environ- beetles identified in this assemblage, such asPtinus ments, and Aphodius luridus, a dung beetle found in fur, it suggests the beginnings of a synanthropic sheep, horse and cow dung in both open and shaded (ecologically associated with humans) element in the (wooded) locations (Jessop 1986). A small number local insect fauna (also vaguely suggested in context of arable/disturbed ground species also occurred [142]), prior to the digging of the town ditch.