Start with museum and national archives

The most reliable starting point is a museum database or national visual arts archive. These hold works that have been catalogued, photographed, and sometimes digitised in reasonable resolution. Search by the artist name and filter by date if you know the period. The results will be incomplete, because many museums still describe works in catalogue records without publishing an image. That gap matters more than it looks. A record without a picture tells you the work exists, but it does not help you judge it. Plan to spend time on records that describe medium, dimensions, and provenance, even when the image is missing.

National archives can be stronger than a single museum search because they aggregate holdings across institutions. The trade-off is that the interface is often less polished and the metadata is uneven. Some records use old catalogue language that is hard to parse. Others list a work under a variant name. Keep a note of any record that looks promising, even if the image is poor. You can often return to it later once you have narrowed the range of dates or subjects you care about.

Use auction records to fill the gaps

Auction archives are useful for the works that museums do not hold or do not show online. They also reveal how often a name appears in the market, which is a rough measure of how much has been documented. Search the artist name across more than one archive if you can. Results will vary by region and by how recently the catalogue was digitised. Older sales may be listed only in summary form. Newer sales are more likely to include an image and a fuller description.

Treat auction records as pointers, not as a complete view. A work that sold once is not necessarily easy to find again, and an image attached to a sale record may be a reproduction rather than the original. The useful signal is consistency. If several records point to the same subject, date range, or medium, you are probably looking at a real cluster of work rather than a stray mention. That cluster is what you should chase next.

Reproduction catalogues are a different record

Reproduction catalogues are worth checking, but they answer a different question. They show which works have been selected for print editions, not which works survive in the historical record. That distinction matters if you are trying to understand the artist rather than buy a print. A catalogue that offers a Raphael Delorme artwork collection in print form may be convenient, but it is a selected slice, not the full body of work. Use it to spot subjects or compositions you had not seen elsewhere.

If you want to browse the complete body of work, do not treat any single reproduction source as definitive. Cross-check its list against museum and auction records. The overlap tells you which works are both documented and commercially available. The gaps tell you where the historical record is thinner than the retail record. That is usually where the more patient research pays off.