A guest blog by Pete Wing

We are opening up the Museum’s Bee type collection to the world by digitising, geo referencing and releasing the specimens online.

The Museum has around 4,500 bee type specimens that date back to the 1700s. These are the example specimens from which the species was named and described. If a researcher wants to identify specimens they have observed, they need to compare the key characteristics that make that species unique.  In the past, this was done by physically travelling to visit the institution that holds the specimen. However, by capturing and releasing high quality images of these specimens online, this can be done remotely wherever a researcher is in the world. We are sharing this resource with the world to increase the numbers who can access the Museum’s collection.

Let’s bee having you

We have been working to digitise the bee (Apoidea) type collection in order to develop a new type specimens digitisation workflow and increase digital access to the Museum’s bee collection. The bee types are kept separate to the rest of the bee collection. This is important as it means that we can digitise the whole of this collection and we are not selecting types out of the main collection which is a much more costly and less efficient way to digitise. The bee type collection has also benifitted from being re-curated and re-housed during this project.

Dorsal, lateral and face views of Xylocopa pulchra Smith, 1854

For this project we have developed a new workflow in which we take three photographs of each specimen. From above the specimen with labels (dorsal habitus) like all our pinned insect workflows. We then take an image from the side (lateral habitus) and the face. Unlike our regular workflow, these images are stacked which means that several images are taken at different focal points through the view of the specimen and then combined. They are also magnified up to 5x to give high definition images which show the specimen and its anatomy in great detail. This detail, combined with the choice of views, is most useful for scientists using this collection online to examine the key characteristics of each species.

Before the Museum closed in March, we had finished the dorsal imaging for all specimens and imaged 30% of lateral and face images. All of these specimens are now available to be viewed through the Museum’s Data Portal.

What’s in a Name?

While we could not continue to image the bee type collection after the Museum closed, we had imaged all the specimens and their labels. We have therefore been able to carry on working on this collection from home.

First we transcribed the data from the specimen labels into a spreadsheet, recording the collection locality and date, collector, sex of the specimen, type status, the collection or donation from which the specimen came and the registration number from when the specimen entered the Museum collection. As this collection has specimens that range from the late 18^th to the early 21st century there are a large number of different styles of handwriting and printed material with varying degrees of legibility. Fortunately, between the digitisers, we have a lot of experience when it comes to reading somewhat illegible scrawl.

Once transcription was complete, we then set about finding the original descriptions for these specimens. This involves a little bit of detective work as the current taxonomic name of the specimen may not be the one it was described under thanks to revisions over the intervening years. The original name, and author, can normally be found by looking at a combination of the specimen record within the Museum’s collection management system and the labels.

However, there is a level of complexity to searching for these descriptions thanks to multiple papers being written by the same author in the same year (some of these scientists were exceedingly prolific – I’m looking at you T.D.A. Cockerell…) and even once found, not everything is available to view remotely as, due to the date of the paper and the journal within which it was published, it may not have been digitised (Biodiveristy Heritage Library is a fantastic resource for many of these). This, therefore, can also mean that it isn’t possible to definitively establish the reference for everything…yet. In these instances we have flagged the records for reviewing later when we are able to return to the museum and access the relevant papers.

There are also some ‘types’ of manuscript names, these are those specimens that have been recognised as ‘new’ by scientists but have not been formally described anywhere. Strictly these names are considered invalid under the International Code of Zoological Nomenclature (ICZN), though they may have been formally described later by a different taxonomist/researcher which can also muddy the waters somewhat. That said, it’s quite good fun and rewarding when you crack something that had had you scratching your head for extended periods.

Having found the appropriate reference and original description we compared the information in the description to that available data on the labels in terms of the sex of the specimen or the primary data of collection. As none of us are Bee experts we use the label data, rather than comparing the description of the specimen itself to the specimen in the image, in order to not make any assertions or assumptions that are incorrect. This practice of ‘type verification’ can be completed later by an expert but enables us to narrow down the specimens that may need more investigation than others.

We’re now just going through some final checks before hopefully passing our data on to an expert to go through the things we aren’t absolutely sure about and any potential issues we’ve raised.

“Bees are rather popular, as insects go, but there is still a huge amount of work to do to describe new species and produce identification keys. If people can’t identify the bees, we can’t associate biological data with a species, i.e. which bee is pollinating which plant? These bee type specimens are much in demand and we have a responsibility to allow the world to see these unique specimens while preserving them for future generations. These old specimens should solve a lot of identification problems and raise awareness of the diversity of these fuzzy vegetarian wasps.”

Gavin Broad, Principal Curator in Charge of Insects

Unlocking the Museum’s collections and making them available to transform the study of natural history and engage the widest possible audience are key to the Museum’s strategic priorities. There are over 1.5 billion natural history specimens in collections around the world. They have the potential to play a critical role in addressing the most important challenge that humans face over the next years: how to map a sustainable future for ourselves and our changing planet. To stay up to date with the Digital Collections Programme’s latest news follow us on Twitter or Instagram. To see the bee types digitised during this project, and over 4.5 million other specimens, visit the Museum’s Data Portal.