Nine of the seventeen makers in this archive never published a serial-to-year table. On those brands a serial number is not a date, and the fastest way to a wrong answer is to treat it as one. What each of them left instead is different, and it is worth knowing which route applies before you start reading marks.
H.H. Scott — revisions, not numbers
Scott never published serial tables; the model years here come from a collector history of the 299 and 222 lines and from period catalogue and price-list literature. Revisions are identified by schematic, not by serial — and the caveat is worth printing: a Scott source states there were "no less than 7 different variations of the 299C, albeit slight differences", and a few very late 299C units shipped with 7868 output valves whose socket layout differs from the 7591. A revision letter is a bracket, not a certificate.
Leak — the label inside the chassis
Leak published no serial tables; the model years come from the brand's own timeline. The practical evidence is the dated label inside the chassis: one documented unit carries a 1961 label. The label and the capacitors are both changed during restoration, so the archive pairs it with the British capacitor codes rather than relying on either alone.
Radford — the series check
Radford published nothing, and the reachable material is thin: the STA-25 is dated 1964–1971 from a specialist archive. Identifying which series a unit belongs to means checking the valve complement, the circuit boards, the wiring and the identification markings together. One caveat above all: the 2011/2013 reissue company builds to the original 1964 drawings, so reissues must not be merged with originals.
Sherwood — the valve version
Sherwood has no decoder and no official model-year archive; the years come from dealer literature and community threads, and the rest is undated here. The route that works is the valve version — with the honest note that one forum thread asks whether the slanted-preamp-valve S-5000 is early or late, and no authoritative answer appears. Where the sources disagree and nobody settles it, the archive reports the disagreement.
Harman Kardon — the model history
No vintage serial chart and no official model-year archive. The Citation II is dated 1959 and the Festival TA230 1958 from the brand's own history page, and the 330 series comes from the same source. Sold figures for the receivers come from a compiled table of 2024 sales.
Dual — model windows and a platter code
Dual model windows come from a published brand history (the 1009 tested in 1964, the 1019 presented in 1965, the 1219 in 1969, the CS 721 in 1976–1979). The platter date code is community-reported, single-source and hedged by its own author — the German wording is ich glaube, "I believe" — so the archive prints it as era context and never as a certificate. Two practical notes: the serial sticker under the platter is frequently lost, and on the 1219 the platter has to come off because the mat is glued on.
What to do on a brand like this
Identify the model, then read the parts: British capacitor codes, EIA stamps on transformers and pots, a dated label, a chassis or platter stamp. The unit is never older than its newest original part, so the component layer gives a floor and the model gives the window. That combination is the answer these six makers allow — see all makers in the archive.