A cybercatalogue of American sand fly types (Diptera, Psychodidae, Phlebotominae) deposited at the Natural History Museum, London

Abstract Background Sand flies (Diptera, Psychodidae, Phlebotominae) are biting flies involved in the transmission of pathogens, including the protozoan parasite Leishmania amongst human and non-human animals (Rangel and Lainson 2009). New information A total of 60 species of American Phlebotominae (Diptera: Psychodidae), distributed amongst 16 genera were studied. A checklist of the primary and secondary type specimens held at the Natural History Museum, London (NHMUK), is given and 968 photographs of the specimens and their labels are made available on a Scratchpads website http://phlebotominaenhmtypes.myspecies.info.


Introduction
There are approximately 1,000 valid described species of sand flies in the world, of which 530 are known to occur in the Americas (Shimabukuro et al. 2017).
The aim of this work is to provide a checklist of Phlebotominae types of the Neotropical and Nearctic regions held in the collections of the Natural History Museum, London (NHMUK formerly BMNH), together with electronic links to, scans of whole slides, plus detail images of the taxonomically informative morphological structures. This work updates the American portion of the 1990 checklist of types deposited at the NHMUK (Townsend and Lane 1990).

History
Owing to the small size and morphological homogeneity of phlebotomine sand flies, few species were described before the turn of the 20 century. Indeed little progress in their taxonomy was made until Adler and Theodor (1926) and Theodor (1932) drew attention to the utility of various internal structures (cibarium & spermathecae) for species differentiation. Currently, taxonomically valid species descriptions of phlebotomine sand flies still rely on adult morphology. Although newer molecular techniques are increasingly used for identifying species (Zapata et al. 2012), they have yet to be formally used as characters in species descriptions. Type specimens are confined to the adult stage, with few taxa described in the immature stages.
Significant donations to the NHMUK collection have come from: (i) Graham Bell Fairchild who worked on the Panamanian fauna in the 1950s and 60s; (ii) a number of early types described by Robert Newstead in the 1920s when he was a Professor at the Liverpool School of Tropical Medicine; (iii) and most significantly, the collection of Oskar Theodor, which the museum purchased in 1981 after his retirement from the University of Jerusalem. The NHMUK has also absorbed the specimen collections of the London School of Tropical Medicine and Hygiene (LSHTM).
The main body of the NHMUK Phlebotominae collection was assembled by Dr D.J. Lewis who, though not a member of staff, was based at the museum for 30 years  whilst in the employ of the Medical Research Council in London. All his material was mounted in Berlese media which is not a permanent mountant. There are many examples of deteriorating Lewis slides in the NHMUK collection ( Fig. 1), as well as instances where Lewis remounted other authors' type specimens into Berlese. His distinctive handwriting and messy slides, often bearing multiple specimens, are instantly recognisable. th

Taxonomic treatment
Phlebotominae taxonomy has a long history of differences in opinion amongst experts and various conflicting classification schemes. The current situation is no exception, with both classification of Young and Duncan (1994), which is based on the previous work of Lewis et al. (1977) and that of Galati (2003b) in use at the same time, for the New World fauna. For a comparison of how these two classifications relate to each other see Fig. 2.
Currently, the NHMUK collections are arranged according to Young and Duncan (1994), but we have used the Galati (2003b) naming systems here to reflect the current knowledge of the group.
In their guide to identification and distribution, Young and Duncan (1994) adopt the classification of Lewis et al. (1977) with only minor changes. They make it clear that their arrangement of taxa does not represent a phylogenetic analysis of the group and, aside from a few comments in the notes, give no information on the relationships between their subgeneric groups. It is clear that their groupings have been arrived at through a phenetic review of morphological characters with little focus on identifying any common ancestry between groups. Lewis et al. (1977) proposed their classification at a time when those working on sand flies were faced with the choice between two family names, several subfamily names and, for some species, three possible generic names. Their aim was to create a stable general framework with flexible subdivisions into which, recent and future ideas on the evolutionary relationships within the group, could be assimilated. They present a review of the early literature on the group and, with reference to the International Code of Zoological Nomenclature (ICZN), proposed the use of the family name Psychodidae Newman 1834 and the subfamily name Phlebotominae Rondani 1840, proposals that have been more or less universally accepted. Their proposal for the use of just five genera (Brumptomyia, Lutzomyia, Phlebotomus, Sergentomyia and Warileya) now seems less successful. At that time, 1977, around 600 sand fly species had been described, with 290 of these in the American genus Lutzomyia (Young and Duncan 1994). The current checklist for American sand flies alone (Shimabukuro et al. 2017) stands at 530 and use of the Lewis et al. (1977) concept of the genus Lutzomyia would place 495 of these taxa into a single genus. It now seems likely that, by placing the emphasis on a stable and simple classification over one that best represented the evolutionary history of the group, Lewis et al. (1977) created a number of polyphyletic groups, particularly their genus Lutzomyia.
The classification of Galati (1995) and Galati (2003a) is more complex with the genus Lutzomyia, as defined in Lewis et al. (1977) and Young and Duncan (1994), split into 23 genera and 22 sub-generic groupings and has been increasingly utilised by authors working on the American fauna.
The study by Galati (1995) represented the first comprehensive attempt to perform a phylogenetic analysis of the American Phlebotominae, comprising a cladistic analysis of 88 adult morphology characters, using the Psychodidae subfamily Bruchomyiinae as the principal outgroup. The classification has since been updated and represents our current a b Figure 2.
a: Comparison of the two classifications. b: Explanation of the symbols used in Figure 2a. Grey boxes (A) on the Galati (2003b) classification indicate novel phylogenetic information provided by the classification of Galati.
Aside from a few remarks given in the text, Young and Duncan (1994) give no information on the phylogenetic relationships between any of the sub-generic groups within their genus Lutzomyia. Red stars (B) indicate new sub-genera or genera in the Galati classification that do not exist in the Young and Duncan classification. Coloured highlighting (C) indicates groupings from Young and Duncan which are not recovered as monophyletic groups in the Galati classification. Solid blue arrows (D) indicate where exactly the same grouping is recovered in both classifications. Broken blue arrows (E) indicate where broadly the same grouping is recovered in both classifications with only very minor changes to the taxa included.
best understanding of the evolutionary relationships within this important group (Galati 2003a, Galati 2003b).
There are some nomenclatural issues with the Galati classification as it stands at the present moment (see Shimabukuro et al. 2017). Within the classification a number of family group and genus group names are attributed to Artemiev from his 1991 work (Artemiev 1991). In this work, Artemiev does not accompany any of his new names with a description that states in words characters that are purported to differentiate the taxa, thus all his new names do not comply with ICZN Article 13.1.1 http://www.iczn.org/iczn/index.jsp and should be considered nomina nuda. Likewise, there are a large number of family group and genus group names in the classification attributed to Galati (1995), where she also fails to accompany any of her new names with a description that can be used to distinguish them. Galati lists the characters used, gives a polarised transformation series and maps the characters and taxa on to a cladogram, but this is not sufficient under the code to render the new names from her 1995 work available. In 2003, Galati updated her 1995 classification (Galati 2003b, Galati 2003a, this 2003 work contains extensive identification keys for all groups and any unavailable names should take authorship and date from this later act of establishment under ICZN articles 50 and 21.

Checklist
The list contains information compiled from the examination of Nearctic and Neotropical species of sand flies from the Diptera Collection at the NHMUK.
Species are listed alphabetically, with genus and subgenus designations according to Shimabukuro et al. 2017. Links to the scratchpad entry and images for each specimen are given in the eventRemarks section of the checklist.
Readers familiar with the classification of Young and Duncan (1994) should refer to the scratchpads website http://phlebotominaenhmtypes.myspecies.info where the specimens have been arranged according to both classifications.
We have excluded from this catalogue two NHMUK specimens bearing "type" labels, for which we can find no record of the given species name in the literature. One labelled as the holotype of Lutzomyia witremundoi (author Feliciangeli), the other labelled as a paratype of L. patiñoi (authors Mangabeira & Galindo). It seems probable that these are unpublished manuscript names and are therefore unavailable under article 8 of the ICZN.

Specimen Images
Photographs were taken in the Sackler Biodiversity Imaging Laboratory (SBIL), at the NHMUK, with an Olympus BX63-CBH microscope equipped with a DP73 camera and differential interference contrast (DIC) illumination. The images were edited in the Cellsens software and stacked images were produced by HeliconFocus 6 software.
Whole slide scans were generated using a SatScan instrument at x5 magnification and images were cropped and processed using the Inselect software, version 0.1.35, available from the Natural History Museum https://naturalhistorymuseum.github.io/inselect.

Data resources Bibliography
Most of the original species descriptions are still under copyright to their original publisher. The bibliographical and historical database of scientific articles on leishmaniases and sand flies hosted by the Laboratoire de Parasitologie at the University of Montpellier is an extremely useful resource http://www.leishpub.univ-montp1.fr.

Specimen Images
The specimen images are available on the NHMUK data portal: https:// doi.org/10.5519/0051226.
The checklist, specimen data and specimen images are available on a Scratchpads website http://phlebotominaenhmtypes.myspecies.info.