Wing morphometrics as a tool in species identification of forensically important blow flies of Thailand

Abstract

Background: Correct species identification of blow flies is a crucial step for understanding their biology, which can be used not only for designing fly control programs, but also to determine the minimum time since death. Identification techniques are usually based on morphological and molecular characters. However, the use of classical morphology requires experienced entomologists for correct identification; while molecular techniques rely on a sound laboratory expertise and remain ambiguous for certain taxa. Landmark-based geometric morphometric analysis of insect wings has been extensively applied in species identification. However, few wing morphometric analyses of blow fly species have been published.

Methods: We applied a landmark-based geometric morphometric analysis of wings for species identification of 12 medically and forensically important blow fly species of Thailand. Nineteen landmarks of each right wing of 372 specimens were digitised. Variation in wing size and wing shape was analysed and evaluated for allometric effects. The latter confirmed the influence of size on the shape differences between species and sexes. Wing shape variation among genera and species were analysed using canonical variates analysis followed by a cross-validation test.

Results: Wing size was not suitable for species discrimination, whereas wing shape can be a useful tool to separate taxa on both, genus and species level depending on the analysed taxa. It appeared to be highly reliable, especially for classifying Chrysomya species, but less robust for a species discrimination in the genera Lucilia and Hemipyrellia. Allometry did not affect species separation but had an impact on sexual shape dimorphism.

Conclusions: A landmark-based geometric morphometric analysis of wings is a useful additional method for species discrimination. It is a simple, reliable and inexpensive method, but it can be time-consuming locating the landmarks for a large scale study and requires non-damaged wings for analysis.

Keywords: Blow fly; Forensic entomology; Species identification; Thailand; Wing morphometry.

Fig. 1

Map of Thailand showing provinces of collection sites for adult blow flies used for wing morphometric analysis

Fig. 2

Right wing of female Ch. chani showing the 19 plotted landmarks based on Hall et al. [28]

Fig. 3

Boxplot showing centroid size of wings for each blow fly species; non-overlapping letters indicate a statistically significant difference (Mann-Whitney U-test, P < 0.0008). Abbreviations: CC, Ch. chani; CM, Ch. megacephala; CN, Ch. nigripes; CP, Ch. pinguis; CR, Ch. rufifacies; CV, Ch. villeneuvi; LC, L. cuprina; LPA, L. papuensis: LPO, L. porphyrina; LS, L. sinensis; HL, He. ligurriens; HP, He. pulchra. Each box shows the median as a vertical line across the middle, the quartiles (25th and 75th percentiles) at its ends, horizontal lines out the box indicate minimum and maximum, and outlier data are plotted as black circles

Fig. 4

Boxplot showing centroid size of wings for each blow fly species (males and females). Only female specimens of Hemipyrellia pulchra were collected, thus it could not be used for classifying between sexes. Asterisks indicate statistically significant difference between males and females (Mann-Whitney U-test, P < 0.05). Abbreviations: CC, Ch. chani; CM, Ch. megacephala; CN, Ch. nigripes; CP, Ch. pinguis; CR, Ch. rufifacies; CV, Ch. villeneuvi; LC, L. cuprina; LPA, L. papuensis: LPO, L. porphyrina; LS, L. sinensis; HL, He. ligurriens. Each box shows the median as a vertical line across the middle, the quartiles (25th and 75th percentiles) at its ends, horizontal lines out the box indicate minimum and maximum, and outlier data are plotted as black circles

Fig. 5

a Scatter plot showing the variation in shape of wings of blow fly genera Chrysomya, Lucilia and Hemipyrellia along the first two canonical variate (CV1 and CV2) axes with 90% confidence ellipses. Each genus was clearly separated from the others. b Transformation grids illustrate the shape changes from overall mean along CV1 and CV2 axes in positive directions. Circles indicate the locations of the landmarks in the mean shape of the sample; sticks indicate the changes in the relative positions of the landmarks

Fig. 6

a Scatter plot showing the variation in shape of wings of 12 blow fly species along the first two canonical variate (CV1 and CV2) axes with 90% confidence ellipses. b Transformation grids illustrate the shape changes from overall average along CV1 and CV2 axes in positive directions. Circles indicate the locations of the landmarks in the mean shape of the sample; sticks indicate the changes in the relative positions of the landmarks

Fig. 7

UPGMA dendrogram showing phenetic relationships of wing morphology among blow fly species constructed based on the Mahalanobis distances between species