# USNM Bumblebee Specimen Images: 73,497 Smithsonian Bombus and Xylocopa Specimens with Darwin Core Occurrence Data

## Facts

- Provider: Smithsonian National Museum of Natural History, Department of Entomology
- License: CC0-1.0 (https://huggingface.co/datasets/MikeTrizna/bees)
- Size: 11 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/MikeTrizna/bees
- Access: public, readable without an account
- Category: Life Sciences
- Source: https://huggingface.co/datasets/MikeTrizna/bees
- Page: https://scigantic.com/archive/242571d4-609a-5148-902d-38c83dd11e59
- Keywords: Life Sciences, Parquet, Hugging Face

## About

Images and occurrence records for 73,497 bumblebee specimens (family Apidae) in the U.S. National Entomological Collection (USNM) at the Smithsonian National Museum of Natural History, one lateral or dorsal image per specimen. Occurrence data follow Darwin Core (species, collection date, locality, collector, identifier, catalog number) and image metadata follow Audiovisual Core, with links to the full-resolution Smithsonian images. Specimens span 1807 to 2020 and are worldwide in scope but mostly North American; 55,301 localities are georeferenced, and 11,421 specimens identified only to genus appear as Bombus sp. or Xylocopa sp. About 49,000 specimens were imaged in a 2014 rapid-capture project and about 25,000 more in 2018; label data were transcribed by volunteers and a professional service. Locality is withheld for the endangered Bombus affinis, B. franklini and B. terricola, and Brazilian specimens have no images. Supports species identification from images, morphology and range studies. Parquet, about 3.7 GB; load with datasets. Public domain (CC0) from the Smithsonian NMNH Department of Entomology; cite NMNH Extant Specimen Records (USNM).

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("MikeTrizna/bees")
print(ds)
```

## Data structure

File types in sampled files: Parquet

Example file: `data/train-00000-of-00008-92b3a17854595f78.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/242571d4-609a-5148-902d-38c83dd11e59
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/242571d4-609a-5148-902d-38c83dd11e59
- Over MCP: `get_schema_card` with id `242571d4-609a-5148-902d-38c83dd11e59` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [Life Sciences datasets](https://scigantic.com/topics/life-sciences)
