New ant species identified by imaging software

Harvard University researchers use 3D imaging system in comprehensiove cataloguing project and confirm existence of a new species of ant from the Dominican Republic

Syncroscopy reports that Auto-Montage, its unique 3D imaging system, has been used at Harvard University to help identify a new ant species from the Dominican Republic.

Researchers at Harvard use Auto-Montage to automatically generate in-depth 3D images of ants, as part of a large, on-going cataloguing project, which aims to produce definitive images of the 13,000 known ant species and help identify new ones.

Since the researchers have to record at least four images of different parts of each ant to identify it, without Auto-Montage software it would be a very time consuming manual task to catalogue species or check potential new ones.

Gary Alpert, an entomologist in the environmental health and safety department at Harvard, explained: "The presence or absence of certain ant species can indicate the ecological damage being done by activities such as mining, deforestation or wars.

"Therefore, it is important to ensure that all species are correctly identified.

"We chose Auto-Montage because it produces close up, accurate 3D ant images that are crucial to species identification.

"The software is so good it has allowed us to compare a digital image of a recently described new ant species, Thaumatomyrmex mandibularis from Cuba, with a species we have just collected from the Dominican Republic.

"This rapid and simultaneous comparison of specimens has enabled us to say with certainty that the ant from the Dominican Republic is indeed a new species".

Martin Smith, Syncroscopy's divisional manager stated: "We are delighted that Auto-Montage has helped entomologists at a well-respected university such as Harvard to identify a new ant species.

"This exciting result will give entomologists world-wide the confidence to use Auto-Montage to generate and apply accurate 3D insect images to their own taxonomic tasks.

Back to top