Geronimo, Chelo Jan S.

TTH 10:00-11:30
BSND-2045602

Mycobiota of cocoa: From farm to chocolate

1. Introduction
Chocolate quality may be influenced by a variety of environmental, agronomic and technological
factors to which the cocoa beans are subjected from the opening of the fruit until the end of
industrial processing. Among these factors, it is known that the microorganisms present in the
fermentation play an essential role in the development of the sensorial characteristics of
chocolate (Rohan, 1964, Schwan and Wheals, 2004 and Afoakwa et al., 2008). The intensity of
microbial proliferation and the presence of one or another group of microorganisms in the
fermentation and drying steps are crucial for the development of particular characteristics in
cocoa beans leading to a final product of good quality (Schwan and Wheals, 2004).
In general the processing of cocoa starts at the farm, where the primary processing is carried out.
Following the opening of freshly harvested pods, the beans are placed in wooden boxes, allowing
natural fermentation for about 6 days. After fermentation, beans are transferred to sun drying
platforms where a gradual reduction in moisture content occurs. When moisture content reaches
values sufficiently low (6–7%), beans are transferred to storage rooms, then later bagged and
marketed. At the manufacturing stage or secondary processing, the dried fermented cocoa beans
are roasted and grinded and then follow two distinctive processing lines giving as final product
powdered cocoa or chocolate (Wood, 1985).
Research has shown a steady and marked presence of fungi during the fermentation process and
during the drying and storage steps (Ribeiro et al., 1986, Ardhana and Fleet, 2003,
Mounjouenpou et al., 2008 and Sanchez-Hervas et al., 2008). When present, filamentous fungi
occur with higher intensity in the last days of fermentation and are usually related to the
formation of off flavors and deterioration (Schwan and Wheals, 2004).
Besides the deteriorative potential and consequent influence on sensorial quality of cocoa and
chocolate, the presence of fungi in food is also a public health issue due to the possibility of
mycotoxin formation. Aspergillus, Penicillium and Fusarium are the main genera capable of
producing toxic secondary metabolites, many of which have mutagenic, teratogenic and
carcinogenic effects on humans and animals ( IARC, 1993). In a previous study, Copetti et al.
(2010) when evaluating whole cocoa beans found the presence of ochratoxin A at all stages of
cocoa processing on the farm, with the major incidence during drying (<0.01–5.54 μg/kg) and
storage (<0.01–4.64 μg/kg). In that study, the capability for production of ochratoxin A by the
isolates of potentially ochratoxigenic species was evaluated, finding out that all Aspergillus
carbonarius and Aspergillus ochraceus group isolates were ochratoxin A producers whereas only
about 5% of Aspergillus niger presented such ability.
Currently, data on mycobiota occurrence during the different periods of the primary processing
of cocoa under natural conditions and on fungal contamination in the products of secondary
processing are limited. Therefore, the aim of this research was to evaluate the mycobiota of
cocoa during the processing steps and in the final chocolate product, correlating the results with
the levels of water activity (aw) in the products.

2. Methods
2.1. Samples
A total of 494 samples of cocoa, corresponding to the stages of primary processing on grower
farms (226 samples), secondary processing at the manufacturing plants (168 samples) and
chocolate sold on the market in the form of powder and bars (100 samples) were evaluated.
The 226 samples of integral cocoa beans (beans and pulp) (not less than 200 g each sample),
from the primary processing stage, were collected in the cocoa producing region of Ilheus-Bahia
(Brazil). The cocoa samples came from three major farms and represented the different stages of
production: 25 samples before fermentation, 51 samples with different times of fermentation (1–
6 days), 85 samples collected at different times of sun drying on movable roof platforms (1–14
days) and 65 samples during the storage period (up to one year).
One hundred and sixty-eight samples from secondary processing were evaluated at the different
stages of cocoa processing. The cocoa samples were collected as follows: 19 testa (shell)
samples, 29 of nibs, 25 of liquor, 25 of cocoa butter, 26 of cake and 44 of powder. These samples
came from three Brazilian manufacturers located in the same region of the farms.
The 100 samples of chocolate were acquired from the market, corresponding to dark (25), milk
(25) and white (25) chocolate in bars; and in powder (25).

2.2. Analysis of water activity

The water activity of samples was directly determined in triplicate in an Aqualab, Series 3TE
instrument (Decagon, USA), at 25 ± 0.1 °C. Each sample was placed in a disposable cup, sealed
in the analytical chamber until vapor equilibrium was reached, and had the dewpoint detected by
infrared.

2.3. Mycobiota of cocoa beans

The cocoa beans were superficially disinfected by immersion in a solution of sodium
hypochlorite 0.4% for 2 min to remove the surface contaminants. Then 11 beans were placed on
3 Petri plates containing agar Dichloran Glycerol 18% with Chloramphenicol (DG18) (pH 5.6),
with a total of 33 beans analyzed per sample. The plates were incubated at 25 °C for 7 days and
the results expressed as a percentage of beans infected internally, according to Pitt and
Hocking’s (2009) methodology.

2
For the samples of industrially processed cocoa, the technique of dilution plating was applied.
Under aseptic conditions, 25 g of each sample were weighed and 225 mL of sterile peptone
water 0.1% were added. Then, aliquots of the serial dilutions were prepared and inoculated on
plates containing DG18. The plates were incubated at 25 °C for 7 days and the results expressed
in colony forming units per gram of sample (CFU/g), according to Pitt and Hocking’s (2009)
methodology.
After incubation, the plates were examined and all the fungal species were first isolated on Petri
plates containing Czapek Yeast Extract Agar (CYA) to be later identified by specific protocols
for each genus.

2.4. Identification of fungi

The isolated Aspergillus sp. and Penicillium sp. were grown in Czapek yeast autolysate (CYA)
agar and Malt extract agar (MEA). The genus Penicillium was identified according to Pitt (2000)
and Samson et al. (2002), and identification of species in the genera Aspergillus and Eurotium
was performed according to Klich and Pitt, 1988 and Samson et al., 2004 and Frisvad et al.
(2004). The other isolates were identified according to descriptions of Pitt and Hocking (2009)
and Samson et al. (2002), supplemented with other sources when necessary.

3. Results
3.1. Fungi and water activity during the primary processing of cocoa
Table 1 shows the values of aw found during the primary processing of cocoa beans at the farm.
The aw values showed a large decrease between the stages, with the biggest variation observed
during the sun drying. The frequency of occurrence and the level of infection by each fungal
species isolated during fermentation, drying and storage of cocoa are presented in Table 1. A
considerable difference in fungal presence, both qualitative and quantitative, was observed
between the processing stages.

Table 1.
Frequency of occurrence, mean and variation in the level of infection of cocoa beans by fungi
during fermentation, sun drying and storage.
Processing stage (number of samples) Fermentation
(51)
Sun Drying
(85)
Storage

3
(65)
Mean aw (range) 0.99 (0.98–0.99)
0.81 (0.49–0.99)
0.65 (0.40–0.85)
Fungi FO (%) AI (%) RI (%) FO (%) AI (%) RI (%) FO (%) AI (%) RI (%)
Absidia corymbifera 15.69 4.94 ND – 97 73.17 26.19 ND – 100 69.23 31.83 ND – 100
Aspergillus candidus 1.96 0.06 ND – 3 20.00 2.19 ND – 27 18.46 1.43 ND – 21
A. carbonarius 1.96 0.06 ND – 3 5.88 1.48 ND – 45 9.23 2.40 ND – 66
A. clavatus 0 ND ND 1.18 0.04 ND – 3 0 ND ND
A. flavus 3.92 0.16 ND – 6 37.64 11.30 ND – 100 32.31 7.74 ND – 100
A. fumigatus 3.92 0.65 ND – 12 11.76 1.69 ND – 66 1.54 0.09 ND – 6
A. niger group 3.92 0.24 ND – 9 16.47 2.65 ND – 60 26.15 4.15 ND – 51
A. ochraceus group 0 ND ND – 0 5.88 0.28 ND – 9 3.08 0.09 ND – 3
A. parasiticus 1.96 0.10 ND – 6 24.39 5.65 ND – 78 13.85 2.22 ND – 48
A. penicillioides 0 ND ND 0 ND ND 15.39 2.20 ND – 60
Aspergillus sp. nov. 3.92 0.24 ND – 9 48.24 18.25 ND – 100 41.54 14.20 ND – 100
A. sydowii 3.92 0.06 ND – 3 2.35 0.35 ND – 24 18.46 2.58 ND – 42
A. ustus ND ND ND 0 ND ND 1.54 0.05 ND – 3
A. versicolor 1.96 0.06 ND – 3 0 ND ND 6.15 1.71 ND – 57
Ascomycetes 0 0.12 ND – 3 1.18 0.04 ND – 6 0 ND ND
Cladosporium sp. 0 ND ND 0 ND ND 1.54 0.05 ND – 3
Dematiaceous hyphomycetes 0 0.06 ND – 3 4.70 3.09 ND – 100 6.15 0.28 ND – 9
Emericella nidulans 0 ND ND 0 ND ND 3.08 0.18 ND – 9
Eurotium amstelodami 5.88 0.84 ND – 27 8.23 0.51 ND – 10 35.38 12.32 ND – 100
E. chevalieri 0 ND ND 12.94 1.13 ND – 27 21.54 2.97 ND – 66
E. rubrum 0 ND ND 9.41 0.92 ND – 18 29.23 7.37 ND – 100
Eupenicillium sp. 0 ND ND 0 ND ND 1.54 0.51 ND – 33
Fusarium solani 0 ND ND 1.18 0.04 ND – 3 0 ND ND
Geotrichum candidum 19.61 9.14 ND – 100 10.59 5.10 ND – 100 6.15 0.78 ND – 21
Monascus ruber 25.49 3.69 ND – 33 4.70 0.56 ND – 24 10.77 1,66 ND – 60

4
Mucor sp. 5.88 0.53 ND -21 1.18 0.11 ND – 9 3.08 0.23 ND – 9
Neosartorya fischeri 0 ND ND 1.18 0.11 ND – 6 1.54 0.09 ND – 6
Paecilomyces variotii 3.92 0.24 ND – 12 4.70 0.41 ND – 21 10.77 0.69 ND – 12
Penicillium citrinum 0 ND ND 8.23 2.96 ND – 39 7.69 1.75 ND – 51
P. fellutanum 0 ND ND 0 ND ND 1.54 0.51 ND – 33
P. paneum 23.53 2.69 ND – 27 58.82 17.31 ND – 100 16.92 3.37 ND – 66
Rhizopus sp. 1.96 0.16 ND – 6 9.41 1.06 ND – 33 10.77 0.74 ND – 12
Syncephalastrum sp. 1.96 0.06 ND – 3 0 ND ND 15.38 2.82 ND – 42
Wallemia sebi 0 ND ND 0 ND ND 1.54 0.32 ND – 21
Yeasts 41.18 28.12 ND – 100 48.23 20.33 ND – 100 9.23 3.42 ND – 69
FO = Frequency of occurrence % (number of samples that contained a fungal species/total of
samples evaluated, %); AI = Average of infection (sum of infection level/total number of
samples); RI = Range of Infection % (Range of infected beans in a sample, %); ND = Not
Detected.

Table options
3.1.1. Fermentation
During the six days of fermentation the fungi most commonly isolated were: yeasts, Monascus
ruber, Penicillium paneum, Geotrichum candidum and Absidia corymbifera ( Table 1). In this
period, the aw of cocoa beans was in the range of 0.99–0.98 (Table 1). In general, filamentous
fungi were found at low levels during this stage.

3.1.2. Drying
At the end of the fermentation the beans were transferred to the sun drying platforms, although
some farms also adopt drying in artificial dryers. In this work the samples correspond to beans
collected exclusively at sun drying, which usually takes 7–14 days or more, depending on
weather conditions. The fungi most frequently isolated during drying were: Abs. corymbifera, P.
paneum, yeasts, Aspergillus sp. nov. (related to A. tamarii), A. flavus, A. parasiticus, A.
candidus, A. niger and Eurotium chevalieri ( Table 1). The new species may be the same as A.
cacao (nomen nudum), mentioned by Raper and Fennell (1965), but this species has never been
validly described.
During this period the aw values declined from about 0.99 found on the first day to around 0.49
at the end of the period (Table 1). This huge variation was also reflected in the diversity of the
fungi isolated. At the beginning of drying, the beans showed a higher aw, so there was a

5
predominance of fungi found at the end of fermentation (Abs. corymbifera, P. paneum and
yeasts). With a reduction in aw the mycobiota started to change to fungal species typically
isolated from intermediate moisture foods, such as the genus Aspergillus and xerophilic fungi
such as A. candidus and E. chevalieri, especially after some of the volatile acetic acid had
evaporated.
In the course of drying, an increased occurrence of Aspergillus section Flavi and Aspergillus
section Nigri was also observed, which are important groups of potentially aflatoxin and
ochratoxin producers, respectively. The prevalence of beans infected with A. flavus reached
100% in some samples. This aflatoxigenic species was isolated from 11.3% of drying samples.
The values found for Aspergillus parasiticus were lower. Regarding the potentially
ochratoxigenic fungi, A. niger was the most prevalent species, followed by A. carbonarius and
the group of A. ochraceus at a lower number ( Table 1). At one stage of the drying process, these
species were exposed to moisture and temperature that may allow mycotoxin production.

3.1.3. Storage
In stored samples, there was an increase in the number of xerophilic species of Eurotium such as
E. amstelodami, E. chevalieri and E. rubrum. Aspergillus penicillioides was primarily isolated at
this stage, reaching up to 100% of infection in some samples. The water activity decreased from
0.85 to 0.40 ( Table 1). A small increase in the diversity of the species Aspergillus and
Penicillium was noticed. The number of yeasts and G. candidum suffered a drastic reduction,
while Abs. corymbifera persisted.

3.2. Fungal contamination and water activity during industrial processing and in the final product
The number of fungi isolated from samples after industrial processing was low (<10–103 CFU/g,
data not shown), mostly corresponding to xerophilic species. Table 2 presents the isolated
species, as well as the number of samples in which they were present.

Table 2.
Fungal contamination in cocoa’s secondary processing products from industry.
Product (a/b) Shell
(19/19) Nibs
16/29) Liquor
(4/25) Butter
(1/25) Cake
(0/26) Powder

6
(11/44)
Abs. corymbifera 14/19 7/29 – – – 1/44
A. candidus – 1/29 – – – 1/44
A. flavus 11/19 8/29 1/25 – – 3/44
A. niger 8/19 8/29 – – – 2/44
Aspergillus sp. nov 7/19 2/29 – – – 1/44
A. sydowii 1/19 1/29 – – – 2/44
A. terreus 3/19 – – – – 1/44
A. versicolor 2/19 2/29 – – – –
Cladosporium sp. – 1/29 – – – –
E. amstelodami 12/19 7/21 – 1/25 – 4/44
E. chevalieri 10/19 11/21 – – – –
E. rubrum 7/19 5/21 – – – –
Em. Nidulans 5/19 7/29 – – – –
Yeast 3/19 – 3/25 –
Paecilomyces sp. 8/19 – – – – 2/44
P. citrinum 3/19 – – – – 2/44
P. paneum 1/19 – – – – –
P. variabile – – – – – 1/44
a/b = number of samples with fungi/total number of samples evaluated.

Table options
The cocoa testa was the industrial product with the highest fungal contamination, basically by
the same species previously isolated from stored beans. Abs. corymbifera, E. amstelodami and
A. flavus were the predominant species.
Despite the heat treatment to which the samples were submitted, 55% of the nib samples
analyzed showed the presence of some fungal species, following a pattern of species similar to
that found in the shell.
The liquor fraction and butter showed low contamination and fungi were not detected in the cake
samples. On the other hand, fungi were detected in approximately one third of the cocoa powder
samples analyzed in this study.

7
No fungi were isolated from the 75 samples of chocolate bars evaluated. Of the 25 samples of
powdered chocolate analyzed, only one showed fungal contamination (104 CFU/g) and the
species isolated were: E. amstelodami, E. rubrum, A. niger, A. flavus, Paecilomyces variotii,
Penicillium citrinum and Aspergillus sp. nov.

Table 3 shows the aw found in the different products. The aw values of industrial and
commercial samples ranged between 0.18 and 0.62.
Table 3.
Water activity values of cocoa products from secondary processing at industry and chocolate
from retail outlets.
Samples Number of samples Water activity (aw)
Mean Variation
Industry 168 0.42 0.18–0.63
Shell 19 0.48 0.32–0.63
Nibs 29 0.47 0.30–0.61
Liquor 25 0.38 0.18–0.62
Butter 25 0.49 0.42–0.55
Cake 26 0.32 0.24–0.48
Powder 44 0.39 0.19–0.59
Market 100 0.44 0.21–0.58
Powdered chocolate 25 0.51 0.43–0.58
Dark chocolate 25 0.41 0.21–0.51
Milk chocolate 25 0.40 0.34–0.54
White chocolate 25 0.44 0.35–0.55
The bold values in the table summarizes the data showed in details in the following lines.

Table options
4. Discussion
4.1. Fungi and water activity during the primary processing of cocoa
This study confirmed the relationship between the availability of water in the food and the ability
of microorganisms to grow.

8
4.1.1. Fermentation
According to Schwan and Wheals (2004) cocoa pulp inside a not injured cocoa pod is
microbiologically sterile but is subsequently contaminated with microorganisms when the fruit is
opened. These microorganisms mainly come from the pod surface, workers’ hands and knives,
insects, baskets used to transport the beans and pulp and contribute to the subsequent
spontaneous fermentation process (Schwan and Wheals, 2004) The role of filamentous fungi in
the microbial succession of cocoa fermentation is not totally known. It is believed that an
extensive fungal development at the end of fermentation causes increased deterioration in the
consecutive phase of drying (Gilmour and Lindblom, 2008). It is known that these
microorganisms can cause hydrolysis of the pulp, produce acids, off flavors and alter the taste of
the cocoa beans (Schwan and Wheals, 2004). Besides this, some fungal species can also produce
mycotoxins in cocoa (Petithuguenin, 2002, Gilmour and Lindblom, 2008, Copetti et al., 2010
and Copetti et al., 2011).
Considering some physiological aspects of the more prevalent fungi isolated at fermentation, Pitt
and Hocking (2009) remark that G. candidum, a yeast-like fungus, grows well under
microaerophilic conditions and high aw and M. ruber and P. paneum are also adapted to
environments of low oxygen pressure and show good growth under these fermentation
conditions. P. paneum is a species closely related to Penicillium roqueforti, which has an
optimum pH for growth between 4.0 and 5.0, and tolerates high levels of CO2 ( Taniwaki et al.,
2001 and Taniwaki et al., 2010), as well as organic acids commonly found in the fermentation of
cocoa (Vivier et al., 1992). Both species were also present in the mycobiota of other foods
fermented by lactic or acetic acid bacteria ( OBrien et al., 2006 and Storm et al., 2008).
The low level of filamentous fungi isolated during cocoa bean fermentation can be explained not
only by the high population of bacteria and yeasts, which have competitive advantage because of
their higher speed of multiplication and metabolites produced during their growth (i.e. alcohol
and organic acids) but also because of the restrictive temperature, as this can rise above 45 °C
after 48 h. However, in our study, one exception was the high prevalence of Abs. corymbifera in
8 samples. This species of zygomycetes is able to grow at temperatures of up to 54 °C ( Pitt and
Hocking, 2009).
A few studies have been carried out on the presence of fungi during cocoa fermentation. Ribeiro
et al. (1986) studying filamentous fungi from cocoa fermentation in Brazil isolated the following
species: Aspergillus fumigatus, A. niger, Fusarium verticillioides, Fusarium oxysporum,
Lasiodiplodia theobromae, Mucor racemosus, Mucor sp., P. variotii, P. citrinum, Penicillium
implicatum, Penicillium spinulosum, Thielaviopsis ethaceticus, Trichoderma viride and Mycelia
sterilia. In our study we found some species described by those researchers; although others were
not isolated. However, the taxonomy of Penicillium and Aspergillus has been significantly
updated the last decade and so has the knowledge on the influence of acetic acid on the fungal

9
mycobiota ( Samson et al., 2010). In the Ribeiro et al. (1986) study the acetic acid tolerant P.
variotii and T. ethaceticus were found, whereas we found other acetic acid tolerant species, such
as M. ruber, P. paneum and Byssochlamys nivea. Likewise we found other species on cocoa with
less acetic acid and lowered water activity. Another difference is the isolation methods and
media used by Ribeiro et al. (1986), which differ from ours.
Mounjouenpou et al. (2008), reported the occurrence of filamentous fungi in an experiment
where cocoa was processed under controlled conditions in Cameroon. Under conditions similar
to those adopted on the farms evaluated in this study, the authors described the presence of
Aspergillus. versicolor, A. fumigatus, A. tamarii, Rhizopus nigricans, Fusarium sp. and A. niger
isolated at the end of the fermentation period. Some of the mentioned species were also isolated
in our study, suggesting the existence of typical mycobiota present in cocoa, even from different
continents.

4.1.2. Drying
The sun drying period has great importance for fungal development. At this stage there is a
gradual reduction in water content of cocoa beans. With a reduction in aw to below 0.88 the
interruption of most bacteria and yeast growth occurs, although there is still enough water to
sustain fungal multiplication (Beuchat, 1987).
In this work, the increase of Aspergillus section Flavi and Aspergillus section Nigri was shown
in this phase of cocoa processing. These are important groups of potentially aflatoxin and
ochratoxin producers, respectively. In studies carried out in Brazil, Copetti et al., 2010 and
Copetti et al., 2011) found a strong correlation between the presence of A. carbonarius and the
level of ochratoxin A contamination in the samples, but only a weak correlation was found
between the occurrence of potentially aflatoxigenic fungi and aflatoxin in the samples.
Mounjouenpou et al. (2008) also reported higher OTA content in cocoa samples where A.
carbonarius had been isolated, in Cameroon.

4.1.3. Storage
The xerophilic species that presented an increase in stored samples were reported as those
responsible for considerable economic losses and common contaminants of stored grains, nuts,
spices and cereal products (Samson et al., 2002 and Pitt and Hocking, 2009). They have little
importance regarding mycotoxin production according to Pitt and Hocking (2009), but long
periods of storage of cocoa seeds should be avoided. Wood (1985) recommended that cocoa
beans in tropical countries can have safe storage only for 2 or 3 months after which there is a
problem of mold development. This happens because cocoa beans are hygroscopic and will
absorb moisture under humid conditions until they reach equilibrium. In adverse conditions, such
as low aw, the fungal spores can remain viable for long periods, even without vegetative growth

10
of hyphae. Storage under poor conditions, in which the cocoa is subjected to environments with
high humidity, can provide suitable conditions for spore germination, fungal growth and decay,
besides the possibility of toxin production. According to the same author, cocoa beans with a
moisture content of 8% or more will turn moldy.
Research by Sanchez-Hervas et al. (2008) evaluated the mycobiota of 9 samples of cocoa stored
from Sierra Leone, Equatorial Guinea and Ecuador and revealed the predominance of
Aspergillus sections Flavi and Nigri. The mycobiota was similar to that found in our study,
although the frequency of occurrence of the species was a little different, especially for
xerophilic fungi.

4.2. Fungal contamination and water activity during industrial processing and in the final product
The low level of fungi in the final product is a result of industrial processing (roasting, grinding,
pressing and alkalizing) which basically consists of technological processes that change the
cocoa beans into products which actually taste like chocolate and at the same time reduces
microbiological contaminants (Beckett, 2008). After pre-roasting of the beans, the testa is almost
completely removed by winnowing (Minifie, 1999), which represents a significant reduction in
the amount of contaminant microorganisms in the subsequent stages (ICMSF, 2005). The high
temperature employed during the roasting and the procedures used to obtain nibs and liquor,
coupled with the remaining moisture in the beans, should kill the microorganisms which can be
present (Beckett, 2008). Treatments from 15 min to 2 h and 105–150 °C in this stage of
chocolate production are capable of completely destroying the vegetative cells of
microorganisms (ICMSF, 2005).
The low contamination found in the liquor and cake was expected due to thermal
treatment/alkalization they were subjected to. In this process nibs, liquor, cake or cocoa powder
are heated with alkali at temperatures of 85–115 °C, which produces a strong sterilizing effect
due to the combined action of water, alkali and temperature (Minifie, 1999). According to this
author, the cocoa powder’s final microbiota is almost exclusively introduced during the
subsequent stages of processing. The few isolated species were similar to those found in stored
beans, suggesting a post-processing contamination. The aw values of industrial and commercial
samples are not conducive to microbial growth. There have been reports of development of
xerophilic fungi such as Bettsia alvei (teleomorph of Chrysosporium farinicola), Chrysosporium
xerophilum and Neosartorya glabra in deteriorated chocolate and chocolate confetti ( ICMSF,
2005 and Kinderlerer, 1997). Usually these episodes were due to inadequate storage problems
arising from the formation of an environment with increased availability of water at the interface
of the packaging and chocolate when it was stored in environments with high relative humidity
(ICMSF, 2005).

11
5. Conclusion
The mycobiota encountered during the processing of cocoa beans were diverse and composed of
at least 20 genera of filamentous fungi and yeasts. Two groups of fungi were predominant: those
that tolerate the antifungal effect of acetic acid, and those that thrive when most of the acetic acid
is evaporated. When there were still substantial amounts of acetic acid and carbon dioxide in the
product, known acetic acid tolerant fungi, such as P. paneum, P. variotii and Monascus were
present, but these species were less common on cocoa beans when the acetic acid disappeared
and the water activity decreased. Regarding the presence of potentially toxigenic species and
levels of water activity required for their multiplication and production of mycotoxins, the data
found in this study suggest the drying stage as critical for mycotoxin production. The occurrence
of xerophilic fungi increased in storage, thus they can be considered potential spoilers at this
stage. With respect to the industrial processing, the water activity values found in processed
products was rather low, not allowing for microbial growth.

12