Institutional databases hold the core material

Most serious research begins in museum and library systems, because that is where documentation survives. Regional collections in the United States and Canada occasionally hold small groups of work by lesser known American painters of the early twentieth century. A search there will usually return a handful of records rather than a full survey. That is normal for an artist whose public profile never reached the scale of a major school.

Use the catalogue fields carefully. A record with a date and a medium tells you more than a thumbnail on a commercial page. Note whether the institution lists a provenance line or simply a credit line. The difference matters when you try to separate an original record from a later reproduction. This is the point where a Francis McMellon art archive stops being a slogan and becomes a set of actual files.

Where to browse McMellon's prints digitally

Commercial sites can be useful for previewing compositions, but they serve a different job from an archive. A reproduction seller shows what a print can look like on paper or canvas, not what the original record contains. Treat those images as a visual reference, and keep them separate from any conclusion about rarity or attribution.

If you want to compare several pieces before deciding, build a short list across two kinds of source. One side gives you institutional documentation. The other gives you a sense of scale and colour in a printed form. The two do not replace each other. The list below separates the kinds of place that actually help.

Here is a practical short list for a first pass.

Search a major museum database for any catalogued holdings and save the record numbers.

Check a regional library digitisation project for scanned letters, exhibition lists, or local press.

Look at one specialist print seller to see which compositions survive as reproductions.

Keep a separate note of any image that appears in more than one place, since that often signals a widely reproduced composition.

What a thin archive actually means

A thin public archive does not mean an artist left no trace. It usually means the trace is uneven. Some work may sit in private hands, some may be uncatalogued, and some may have been reproduced under a variant name. That is the realistic shape of the record for a minor figure, and it is why a single definitive site rarely exists.

The practical response is to lower the expectation of completeness and raise the expectation of cross checking. Compare a dated museum record with a reproduction you can actually buy. If the two do not line up, that is information. If you want to understand the difference between a reproduction and an original record, the question is less about access than about what each source is designed to show.