Castillo

Castillo is a coffee variety bred in Colombia by Cenicafé, the research centre of the country’s growers’ federation, and released for commercial planting in 2005.¹² It belongs to the Catimor family — Caturra crossed with the Timor Hybrid, the natural Arabica–Robusta cross that carries resistance to Coffee leaf rust.¹²

It is the most consequential variety on this wiki that almost nobody outside Colombia asks for by name. From 2008 a government-backed programme replaced billions of Caturra trees with Castillo, and today it is somewhere between 40% and 45% of everything Colombia grows.¹³ Nothing else in coffee comes close as an example of a single disease deciding what an entire country plants.

Two things make it more interesting than a standard rust-resistant workhorse. It is not one plant — “Castillo” is an umbrella over seven regionally selected versions.⁴⁵ And unlike most of the Catimor family, the case that it ruined Colombia’s cup has been tested and largely lost: a controlled trial cupped Castillo and Caturra side by side off the same farm and found similar scores, with some Castillo samples over 90 points.¹

Where it came from

Cenicafé’s genetic-improvement programme started in 1968.¹ Its first product was Variedad Colombia — the country’s first rust-resistant release, also built on Caturra × Timor Hybrid. Castillo is the next generation of that same breeding line, selected over many more cycles and released in 2005.¹²

The name is a tribute. The variety is named after Jaime Castillo Zapata, the Cenicafé researcher behind the rust-resistant Variedad Colombia work.²³ Colombia is unusual in naming a national cultivar after the scientist rather than the region or the institute.

The method matters for understanding what Castillo is. Rather than fixing one line and shipping it, Cenicafé took the Caturra × Timor Hybrid cross through multiple generations of selection followed by five years of field evaluation at different locations, and kept the lines that performed best in each place.⁴ That is why the release is a composite rather than a clone.

To verify — the timeline does not quite add up, and 1983 is doing two jobs

• The dates inside one source conflict. Perfect Daily Grind states both that the programme “began in 1968” and that Castillo was released in 2005 “after 23 years of research and development.”¹ 1968 to 2005 is 37 years; 23 years back from 2005 is 1982. The likely reading is that 23 years counts from the start of Castillo-specific selection, not from the start of the programme — but the article does not say so.
• What happened in 1983. French Wikipedia says Jaime Castillo Zapata “discovered in 1983 the rust-resistant Colombia variety.”² This wiki’s Colombia note records 1983 as the year Coffee leaf rust arrived in Colombia. Those are compatible (resistant material identified as the disease landed) but they are not the same claim, and no Cenicafé document confirming either date was read here.
• No primary release notice was found. 2005 is consistent across every source consulted,¹²⁶⁷ but no publicly reachable Cenicafé bulletin listing “Castillo — released 2005” as a formal line item was located. Cenicafé’s own digital repository is online and searchable⁸ and is the place to settle it.

The variety catalogue has no entry for it

This is worth stating plainly, because it changes how every number on this page should be read. The World Coffee Research variety catalogue — the reference this wiki leans on for Lempira, IHCAFE 90, Anacafe 14 and every other Catimor line — has no retrievable record for Castillo. A request for /varieties/castillo returns 404.⁹ So does /varieties/colombia, for its parent variety.⁹

The consequence: for Castillo there is no independent rating of yield potential, stature, bean size, nutrition requirement, planting density or quality potential at altitude of the kind that anchors the other variety notes here. Everything agronomic below comes from trade and roaster sources, and should be read one tier down accordingly.

That absence is also the answer to a question the Caturra note has been carrying: whether Caturra is really a Castillo parent. The catalogue never said. Two sources read directly for this note — a trade publication citing Cenicafé and French Wikipedia summarising Cenicafé’s work — both state Caturra × Timor Hybrid independently,¹² and Cenicafé’s own technical pamphlet is reported to say the same of both Colombia and Castillo.⁴ Treat the parentage as established; treat the catalogue as silent.

Not one variety — the seven regional Castillos

“Castillo” behaves less like Geisha and more like Catimor or Sarchimor: a name covering several distinct things. Cenicafé released seven regional versions, each selected at and named for the regional experiment station where it was evaluated.²⁵

Regional line	Named for
Castillo Naranjal	Naranjal experiment station, Caldas
Castillo El Rosario	El Rosario station
Castillo La Trinidad	La Trinidad station
Castillo Pueblo Bello	Pueblo Bello station
Castillo Santa Bárbara	Santa Bárbara station
Castillo El Tambo	El Tambo station
Castillo Paraguaicito	Paraguaicito station

The seven-line list is confirmed by two independent sources.²⁵ The lines differ in their rust-resistance genes and in what conditions they were adapted to, which is the point — a variety deployed nationwide across wildly different altitudes and rainfall needs more than one answer.⁴

Practically, this means “Castillo” on a bag tells you less than it looks like it does. It is a pedigree and a breeding programme, not a single plant with a single flavour.

To verify — the station names

The seven line names are solid.²⁵ The station locations in the right-hand column above are the naming convention as described by the sources, not a Cenicafé station register read here — only Naranjal is well documented elsewhere as a Cenicafé experiment station. Do not cite the specific departments for the other six without checking Cenicafé.⁸

Partial cross-check since: the Cenicafé note now records an eight-station register from an encyclopedia source, and five of the seven line names above — Naranjal, El Rosario, Pueblo Bello, El Tambo, Paraguaicito — match stations in it. La Trinidad and Santa Bárbara match nothing in that register. Still no Cenicafé document behind either list.

How Colombia became a Castillo country

Coffee leaf rust reached Colombia in 1983, and Caturra — rated very highly susceptible — was indefensible. It was also, at that point, close to half of national production. From 2008 a government-backed replanting programme swapped billions of Caturra trees for Castillo.¹⁰

The reasoning was never about flavour. A resistant tree that cups slightly less brilliantly beats a susceptible tree that dies, and rust was going to take the crop either way. The full story sits on the Caturra and Colombia notes; the short version is that Colombia’s national average tree is now a Cenicafé hybrid, while the Cup of Excellence and specialty tables still see plenty of Caturra, Bourbon and Geisha from the same country.

To verify — 40% or 45%?

Sources disagree on Castillo’s share of Colombian plantings, and neither figure traces to an official census.

• 45%, attributed directly to Cenicafé by Perfect Daily Grind (2017).¹
• ~40%, given by MTPak’s varietal guide (2022) as sector analysis.³ Other sources decline to put a number on it and simply call Castillo the country’s most planted variety.⁷¹⁰ Quote “roughly 40–45%” or “the most planted variety in Colombia.” A precise share needs an FNC or Cenicafé statistic, which was not located.

The quality argument, and how it was settled

Castillo arrived carrying the Catimor family’s reputation problem: rust resistance borrowed from Robusta via the Timor Hybrid, and therefore — the assumption went — a worse cup. Specialty buyers treated it as a downgrade on principle.¹³

Then somebody ran the experiment. The Colombia Sensory Trial of 2014–2015 compared Castillo and Caturra grown on the same farm and processed identically, which is the only way to isolate the variety from everything else that moves a score. The result: flavour profiles varied a great deal, but the two varieties returned similar cup scores — and some Castillo samples cupped at 90+ points.¹

That is a genuinely important finding and it is why Castillo deserves a fairer hearing than “Catimor” usually gets. Note carefully what it does and does not say. It does not say Castillo tastes like Caturra — the profiles differed. It says that at the same site, with the same processing, the variety was not the thing holding the score down. Lot-to-lot variation swamps the varietal difference.⁷

The general descriptors reported for Castillo cups are smooth body, balanced to citric acidity, red fruit and caramel, usually roasted light to medium.³⁷ Those are trade impressions from cupping tables rather than panel measurements — read them as such, and see Coffee scoring and Coffee Value Assessment for how a claim like this would actually have to be settled.

To verify — every specific number attached to a Castillo cup

• The 90+ figure. The trial claim above is single-sourced to a 2017 trade article citing the Colombia Sensory Trial.¹ The trial’s own published report was not located, so the sample count, the protocol, the panel and the exact scores are all unverified here. MTPak separately says Castillo has “earned 90 points or more during cupping evaluations” without naming a cupping.³ Same claim, no independent paper trail.
• A “Q-grade 90” lot from Planadas, Tolima, at 1,950 m, is claimed in a roaster’s product video.¹¹ That is marketing, not a Q Grader report, and no score sheet is public. Do not repeat the number as a graded score.
• No auction record. No Cup of Excellence, ACE or other official auction result listing a Castillo lot with a lot ID, score, price or buyer was found in anything consulted for this note. Statements about named competitions, prices per pound or buyers for Castillo lots are currently unsupported. The Alliance for Coffee Excellence results archive is where to look.

What the plant is like on the farm

With the catalogue silent, this section rests on trade sources — treat it as the trade’s working description, not a rating.

Castillo is a compact, dwarf-statured tree in the Caturra mould, plantable at high density and high yielding — the whole reason the cross was made on a Caturra frame.⁴⁵⁷ One producer-facing source describes it as slightly taller than Caturra, with long branches, large leaves and large beans.¹⁰ Bean size is put at medium to large, and optimal altitude at 1,200–2,000 m.⁷

To verify — the altitude band and the yield claim

The 1,200–2,000 m range is single-sourced to one roaster’s variety sheet.⁷ It is also, note, the identical band that this wiki’s Caturra note already flags as a single-source claim for Caturra — the two may simply be repeating one another. With no World Coffee Research entry to check against,⁹ no altitude band should be quoted as established for Castillo. The same applies to “high yield”: plausible for a Catimor bred for productivity, and universally repeated,⁴⁵⁷¹⁰ but with no independent rating behind it.

On processing there is nothing variety-specific to report. Castillo is run through the full Colombian range — washed overwhelmingly, and honey and natural where the farm and buyer want it. Sources implying “Castillo is typically washed” are describing Colombian norms, not the variety.

Is the resistance holding?

This is the question that decides Castillo’s next twenty years, and the honest answer is mostly, so far, with maintenance.

The positive record: twelve years after release, Castillo was still reported resistant to both leaf rust and coffee berry disease, with researchers hoping to hold that for another fifteen.¹ Twelve years of field durability is a real result — compare Lempira and Costa Rica 95, Catimor lines rated intermediate that were later confirmed susceptible in their own countries.

The caution: rust is a living organism under selection pressure, and Castillo’s resistance is not one gene but different gene combinations across the seven regional lines.⁴ Breeding has not stopped.

To verify — the "nearly 40 improved versions" claim

Sweet Maria’s glossary is reported to say that rust outbreaks have still occurred in Castillo plantings, prompting Cenicafé to develop “nearly 40 improved versions of Castillo.”¹² That page returned HTTP 403 to automated checks here and could not be read, no second source repeats the figure, and it is hard to square with the seven documented regional lines unless it counts internal breeding lines rather than releases. Do not repeat “40 versions” as fact.

Ownership and the ® mark

Cenicafé’s own technical material refers to “Variedad Castillo®”, which indicates a registered mark held in connection with Cenicafé and the Federación Nacional de Cafeteros.⁴

Beyond that, nothing was verified. No plant breeders’ rights number, patent code or registration identifier was located in anything read here, and no claim about the legal form of Castillo’s protection should be made. What is clear is the practical arrangement: Castillo was bred by a growers’-federation research centre for that federation’s own members, and distributed to them — the opposite of a proprietary variety licensed out to buyers.

Why it matters

Castillo is the Catimor family’s biggest real-world test, and the Anacafe 14 story in reverse. Anacafe 14 is the small, clever fix — cross back to Pacamara, recover the cup, release it to a few thousand farms. Castillo is the blunt instrument: take the standard Caturra × Timor Hybrid cross, select it hard for a decade, split it into seven regional versions, and put it under an entire national industry.

It also quietly demolishes the most common piece of specialty snobbery about the family. The assumption is that Robusta ancestry means a worse coffee; the one controlled same-farm comparison anyone ran found the variety was not what was capping the score.¹ What Castillo actually costs is legibility — a name that covers seven plants and 40-odd percent of a country tells a buyer almost nothing, which is precisely why Colombian lots get sold on farm, region and altitude instead.

And the last thing worth noticing is the gap in the record. The most planted variety in the world’s third-largest coffee origin has no entry in the reference catalogue of coffee varieties.⁹ Documentation follows prestige, not acreage.

Related

• Catimor — the family it belongs to, and the line-by-line comparison it is missing from
• Caturra · Timor Hybrid — its two parents: the productive dwarf, and the borrowed resistance
• Robusta · Arabica vs Robusta — the second species behind that resistance, and why the ancestry is controversial
• Colombia — the origin it now defines, and the federation that bred it
• Coffee leaf rust — the disease that put it in the ground
• Anacafe 14 · Lempira · IHCAFE 90 · Sarchimor — the sibling rust-resistant lines, and how differently they turned out
• World Coffee Research — keeper of the catalogue that has no record of it
• Cup of Excellence · Alliance for Coffee Excellence · Green coffee auctions — where a Castillo competition record would show up if one existed
• Cupping · Coffee scoring · Coffee Value Assessment · Q Grader — how the 90-point claims would have to be settled
• Washed process — the method most Castillo is processed by, and whose textbook definition also comes from Cenicafé
• Climate change and coffee — the next round of the same borrowed-resistance bet
• Specialty coffee — the market that spent a decade assuming this variety was a downgrade

Official links

• Official site — Cenicafé digital repository (Centro Nacional de Investigaciones de Café “Pedro Uribe Mejía”)
• Official site — Federación Nacional de Cafeteros de Colombia

Footnotes

1. Perfect Daily Grind — Coffee Varieties: Debunking The Myths Around Castillo (June 2017) — confirmed via WebFetch; source of the 1968 programme start, 2005 release, 45% share attributed to Cenicafé, the 2014–2015 Colombia Sensory Trial and the 12-year resistance record ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵

2. Castillo (café) — Wikipédia (French) — confirmed via WebFetch; source of the Jaime Castillo Zapata eponym, the 1983 date, the Caturra × Timor Hybrid parentage, the 2005 launch and the seven regional lines ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰

3. MTPak Coffee — Coffee varieties: a roaster’s guide to Castillo (March 2022) — not read directly here ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

4. Cenicafé — Variedad Castillo®: Café Resistente y Productivo (document copy) — Cenicafé technical pamphlet, not read directly here; claims carried from the research report behind this note ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

5. Coffee & Tea Culture — Castillo coffee variety: Colombia’s rust fighter — not read directly here ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

6. CRS Coffeelands — Where did Castillo come from? — unconfirmed: returned HTTP 403 to automated checks here ↩

7. Volcana Coffee — Castillo coffee variety: origins, agronomics & cup profile — roaster variety sheet, not read directly here ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

8. Cenicafé — digital repository, Centro Nacional de Investigaciones de Café “Pedro Uribe Mejía” — confirmed via WebFetch as Cenicafé’s official publications repository; no Castillo record was surfaced from the landing page ↩ ↩²

9. World Coffee Research variety catalogue — confirmed via WebFetch: /varieties/castillo and /varieties/colombia both return HTTP 404 ↩ ↩² ↩³ ↩⁴

10. The Colombian Coffee Co. — Castillo — not read directly here ↩ ↩² ↩³ ↩⁴

11. What is Castillo? Single origin varietal coffee explained (roaster video) — marketing tier; not viewed here ↩

12. Sweet Maria’s coffee library — Castillo glossary entry — unconfirmed: returned HTTP 403 to automated checks here ↩