The ten genetic groups of cacao

What became of the old names

Criollo

Kept as the name of one of the ten groups. It was always a traditional cultivar, and the genetics agree that it is one distinct, very uniform population. [1]

Forastero

Not a group. It meant everything that was not Criollo, and covers what are now nine separate, strongly differentiated populations. The paper proposes its ten groups in place of the old three names. [1]

Trinitario

Not a group either, but hybrids: of Criollo with Amelonado, and of those hybrids with other Amazonian trees. The 2008 study excluded them on purpose, as mixtures made by people. [1]

Nacional and Amelonado

Two traditional cultivars that the old scheme filed under Forastero. Each turned out to be a group of its own, and both names were kept. [1]

Trinitario’s parentage was established separately: a study of Trinitario trees from 17 countries concluded that they result from hybridisation between Criollo and Amelonado. [7] The catalogue’s own entries for Forastero and Trinitario treat them as the trade names they are.

Five clone codes, and where they sit

Clone names such as SCA-6 or SCA-12 begin with letters that say where a tree was collected. Five of the best-known series come from a single upper-Amazon collection, made by F. J. Pound, and the paper places each in a group. [1]

The last column matters more than it looks. Nine in ten of the Nanay trees in the study, and two in three of the Marañón ones, descend from that one expedition, in which, on one reading of the collector’s report, the seed for each series came from somewhere between one and twenty mother trees. The paper notes that this collection is the material most used by breeding programmes worldwide, so much of what breeders know as “upper Amazon” cacao is a narrow sample of it.

For how clone names work in general, see how to read cacao clone codes. Any individual clone in the study can be looked up below.

Look up a clone

The study published the full list of the 952 trees it kept, with the group each was placed in. The article is in the public domain, so the list is here to search and to download as a CSV file.

A match is a match on the name. The study set aside 23% of the trees it sampled because their labels could not be trusted, so a tree sold under a name in the table is not thereby the tree that was genotyped.

Not in the table: CCN-51

The clone most often asked about is absent, because the study left out hybrids. What is on record about its parentage comes from a peer-reviewed review.

• CCN-51 is traditionally recorded as a three-way cross, (IMC-67 × ICS-95) × a tree known as Canelo or Oriente 1, made by Homero Castro Zurita on his farm in the Naranjal area of Guayas province, Ecuador, in response to witches' broom disease. [6]
• The review reports a 2014 genotyping study by Boza and colleagues as confirming IMC 67 and ICS 95 as parents, with IMC 67 the larger contributor. The third parent, Oriente 1, was not found in any germplasm collection, so nothing genetic could be tied to it. [6]
• The same work is reported as finding CCN-51 highly heterozygous and almost unrelated to Nacional: only about 1% of the alleles analysed were associated with Nacional ones. [6]

The catalogue’s entry is CCN-51.

How the ten were arrived at

• 1,241 trees held in germplasm collections, originally gathered in 12 countries between 1937 and 2005, were typed at 106 genetic markers, of which 96 were kept.
• 289 of them, 23%, were set aside before analysis as mislabelled, duplicated or hybrid. 100 pairs of supposedly different trees were genetically identical. The paper also cites an earlier estimate, made from collection records, that between 15% and 44% of trees in cacao germplasm collections are misidentified.
• Of the 952 that remained, 416 came from Peru, 248 from Brazil and 172 from Ecuador.
• 735 belonged clearly to one group. The rest were mixtures, and more genetically varied, which the authors read as trees from zones where populations meet.
• Differences between the ten groups accounted for 38.1% of all the genetic variation measured, and the groups divide further into 36 subgroups.

The greatest diversity was in the upper Amazon of Peru and Brazil, in keeping with the species having originated there. The genome study later found diversity falling steadily from the Pacific side of the basin to the Atlantic side. [1] [2]

Where the trees are kept

The trees in these studies came from living collections: fields of cacao trees, each propagated from a named original. The studies were possible because those collections exist, and their limits are the collections’ limits.

• More than 40 collections around the world hold over 24,370 accessions of cacao between them, with a good deal of duplication within and between collections. [4]
• Two are international collections: the Cocoa Research Centre of the University of the West Indies in Trinidad, and CATIE in Costa Rica. Both have agreements under the international plant treaty to keep their collections for the long term and to supply material freely to any qualified institution or person. [4]
• Outside those two there is little movement of cacao between countries. Plants that do travel go through quarantine, chiefly at the International Cocoa Quarantine Centre at the University of Reading, where they are tested for viruses. [4]
• The Trinidad genebank was begun in 1982 on a site of about 100 acres at Centeno, by bringing together earlier collections: the 100 Imperial College Selections, the ICS clones, from F. J. Pound's survey of Trinidad and Tobago in the 1930s; selections from Marper Farm, which Pound set up after his upper-Amazon expeditions of 1937 to 1942; and other national collections. By 1994 over 2,000 different types had been planted. [5]
• The network lists 32 institutes: the two international collections, the quarantine centre, and 29 national collections in 21 countries and territories, from Brazil and Peru to Ghana, Indonesia and Papua New Guinea. [4]
• Mislabelling is acknowledged by the network itself, which puts misidentification of trees within collections at as high as 30%. Few collections keep a safety duplicate of their unique material at a second site. [4]
• Although scientific breeding of cacao began more than 70 years before the strategy was written, only about a quarter of cacao farms were then planted with improved varieties. [4]

The counts are from a strategy written in 2012 and will have changed. The database that catalogues individual accessions, the International Cocoa Germplasm Database, could not be reached when this page was written, so no collection’s holdings are listed.

What this does not tell you