harbour yeasts that could potentially initiate cocoa bean fermentations.
This study documents cocoa bean fermentation data from east Cuba,
from where such data were missing.
2. Materials andmethods
2.1. Harvest and fermentation
Field experimentswere carried out in 2013 and 2014. Selected cocoa
podswere cut fromthe trees using a 2 to 3mlong stickwith an attached
knife. The podswere piled up on the ground close to the harvest site. On
the delivery day, pods were opened with a knife and the cocoa pulpbean
mass scooped out from the pods by hand, sometimes protected
by a glove, into a yagua. Fabric bags were filled with 40 to 50 kg of
cocoa pulp-bean mass each from the yagua and transported by horse
to either a collection point or the cooperative. The duration of the transport
and waiting period in bags varied between 8 h to two days. At the
cooperative, cocoa pulp-bean mass from different farms was used to
form long fermentation heaps (several m) of about 70 cm height on a
concrete platform exposed to the sun (further referred to as industrial-
scale fermentations). The heaps were covered by a black plastic covering.
Fromday 3 onwards, the heaps were turned daily in the morning
by manual shovelling. Fermentation was considered as completed
when the inner colour of the beanswas reddish,which usually required
seven days. Also, two small fermentations, further referred to as domestic-
scale fermentations (not turned), performed in yaguas close to the
respective harvest sites were sampled.
2.2. Sampling sites
The sampling sites and their specifications are listed in Table 1 and
visualised in the Supplementary Fig. 1. Cocoa-related sites locatedwithin
a radius of 3.5 km were contrasted with rural sites without cocoa
trees or cocoa bean fermentation activities. These rural siteswere located
within a radius of about 40 km.
2.3. Fermentation sampling
In the case of the industrial-scale fermentations, the cocoa pulpbean
mass was transported on the harvest day (2013 field experiment)
orwithin three days after the harvest (2014 field experiment) to the cooperative
and was set up into large fermentation heaps together with
cocoa pulp-bean mass from other plantations. The domestic-scale fermentation
heap was formed directly after harvest. Four industrialscale
fermentation heaps aswell as one small domestic-scale fermentation
(Table 2) were followed over their entire duration by three to four
samples on day 1 and 2, two samples on day 3, and then daily samples
until yeast growth ceased (between 80 and166 h) or until the beans
were spread out for drying. Two fermentations were only partially
sampled.
2.4. Sampling methods
Sampling of cocoa pulp-bean mass was started 10–20 min after
opening the cocoa pods. Initial samples were taken from the yagua,
followed by samples from the fabric bags used by the farmers to transport
the cocoa pulp-beanmass, and finally fromthe assembled fermentation
heaps towhich the content of the transport bags had been added.
One sample of 50 to 100 g from about 20–30 cm below the surface was
collected in sterile plastic sample bags per sampling site and time point.
The sampleswere transported to the laboratory on ice and processed on
the same day or kept refrigerated for one night. Flowers and tree exudateswere
either collected in sterile tubes and later submerged in saline
(0.85% NaCl, m/v) or liquid was taken from the flowers by 10 μL
microcapillaries and transferred into tubes containing saline. Insects
were either enclosed in tubes and later submerged in saline or enclosed