Materia medica databases, compared
A materia medica database is the remedy literature held as queryable data rather than as books: hundreds of remedy entries across multiple authors, indexed so a reader can ask one question of the whole corpus at once. Vendors apply the phrase loosely — anything from a folder of digitised texts to a structured, source-attributed search platform — and the difference determines what questions the database can actually answer.
Two database shapes
Full-text collections store books as books: each text searchable individually, with no shared index. Médi-T's hypertext classics and the Internet Archive's scans are the public examples. They answer "what does Boericke say about Sepia?" perfectly and "which authors connect Sepia to indifference?" not at all.
Structured corpora index every passage by author, source, edition, and remedy, and search the whole set through one query layer. Similia, RadarOpus, Complete Dynamics, and Zomeo are all structured corpora of different depths. They answer cross-corpus questions, filter by author, and link materia medica passages to repertory rubrics.
The comparison
| Dimension | Médi-T / Archive.org | Similia | RadarOpus | Complete Dynamics |
|---|---|---|---|---|
| Shape | Full-text collection | Structured corpus | Structured corpus | Structured corpus |
| Coverage | Public-domain classics only | Classic library free; Murphy, Saine-era and author libraries on Pro or one-time purchase | Largest licensed catalogue in the category | Core classics; more by edition |
| Cross-corpus search | None | Keyword and semantic, with author filtering and priority | Keyword and proximity across libraries | Keyword with filters |
| Source attribution | Per book | Per passage, with author grouping | Per passage | Per passage |
| Cost of entry | Free, no account | Free plan; Pro from $19.99 USD/mo | Paid desktop licence | Free browser tier; paid editions |
| Repertory link | None | Same workspace as the classic repertory | Tight suite integration | Tight suite integration |
What to evaluate
Coverage versus retrieval. The instinct is to maximise the number of texts; the working reality is that retrieval quality dominates once coverage passes the classical core. A database holding twelve sources you can interrogate precisely outperforms sixty behind literal-string search, because classical vocabulary defeats literal matching — "dread of solitude" versus "fear of being alone" is the standard example. Semantic retrieval, where the query matches meaning rather than spelling, is the single highest-leverage feature in the category.
Licensing reality. Everything pre-1930 is public domain and appears in every database; the modern interpretive corpus — Murphy, Vermeulen, Mangialavori, Scholten — is licensed per vendor. A practitioner attached to a specific modern author should confirm that author's presence in the catalogue before committing.
Attribution discipline. A database that returns passages without author, source, and edition strips the metadata that makes the literature usable. The classical works disagree with each other by design — Hering's verification grades and Kent's interpretive portraits are different kinds of claims — and a reader who cannot see which author said what loses the ability to weigh them.
Exit cost. Ask what leaves with you: notes, cases, annotations. The corpus itself is largely public domain; your work product is not, and vendor formats differ in how gracefully they let it go.
Which database for which reader
The student with a reading list needs no database at all — the full-text collections cover assigned classical reading free. The self-studier working cases needs a structured corpus with a repertory in the same workspace; free tiers make that costless to trial, and running a phrase from a real case against a semantic materia medica search is the fastest way to feel the difference between literal and meaning-based retrieval. The established practitioner with a licensed desktop suite already owns a deep structured corpus and should evaluate whether semantic retrieval justifies adding a second tool rather than replacing the first.
References
Boericke, W. (1927) Pocket Manual of Homoeopathic Materia Medica with Repertory, ninth edition, Philadelphia: Boericke & Tafel.
Complete Dynamics (2026) Editions and features, vendor page, https://www.completedynamics.com/, fetched 2026-05-17.
Hering, C. (1879–1891) The Guiding Symptoms of our Materia Medica, ten volumes, Philadelphia: F. E. Boericke.
Internet Archive (2026) Texts collection — homeopathic materia medica scans, https://archive.org/, fetched 2026-05-17.
Médi-T (1998–2026) Homéopathe International — hypertext materia medica editions, http://www.homeoint.org/, fetched 2026-05-17.
Mind Technologies (2026) Hompath Zomeo — features, vendor page, https://www.hompath.com/, fetched 2026-05-17.
Similia (2026) Knowledge base — Free vs Pro features, Materia Medica search modes, author priority, premium libraries and pricing, https://similia.crisp.help/, fetched 2026-05-17.
Zeus Soft (2026) RadarOpus — libraries and packages, vendor page, https://www.radaropus.com/, fetched 2026-05-17.
Verdict
Ready to act on this?