Content uploaded by Andreas Buerkert
Author content
All content in this area was uploaded by Andreas Buerkert on Apr 30, 2018
Content may be subject to copyright.
The role of “effective microorganisms” in the composting of banana
(Musa ssp.) residues
Beate Formowitz1, Fritz Elango2, Shuichi Okumoto2, Torsten Müller3,and Andreas Buerkert1*
1Organic Plant Production and Agroecosystems Research in the Tropics and Subtropics, Institute of Crop Science,
University of Kassel, 37213 Witzenhausen, Germany
2EARTH University, Apdo. 4442–1000, San José, Costa Rica
3Institute of Plant Nutrition, University of Hohenheim (330), 70593 Stuttgart, Germany.
Accepted July 23, 2007
Abstract
“Effective microorganisms” (EM) are a poorly defined mixture
of supposedly beneficial microorganisms that are claimed to
enhance microbial turnover in compost and soil. In Costa
Rica, EM are used to produce organic compost (bokashi)
from banana residues (Musa ssp.). Given the scarcity of
scientific data about the effects of EM on the mineralization of
plant residues, this study aimed at investigating the effects of
EM addition on the decomposition of banana residues during
Bokashi production. To this end, the following non-EM treat-
ments were compared to EM Bokashi: Bokashi produced
with water (W), with molasses (M) as an EM additive, and
with sterilized EM (EMst). Subsequently, the effects of the
resulting Bokashi treatments on the growth of young banana
plants were evaluated. Compared with non-EM controls, the
effect of EM on the mineralization of banana material was
negligible. Dry-matter losses of the composts with different
EM treatments were similar, with about 78% over 5 weeks.
Ergosterol concentration was highest in EM Bokashi
(77 lg (g dry soil)–1) and lowest in EMst Bokashi (29 lg (g dry
soil)–1). Microbial biomass carbon (Cmic) and microbial
biomass nitrogen (Nmic) were both lowest in EM (Cmic =
3121 lgg
–1;N
mic = 449 lgg
–1), while Cmic was highest in
Bokashi produced with molasses (3892 lgg
–1) and Nmic was
highest in EMst (615 lgg
–1). Treatment effects on adenylate
concentrations and adenylate energy charge were negligible.
Application of all Bokashi variants to young banana plants
significantly increased shoot growth under greenhouse condi-
tions compared to plants grown in a control soil without
amendments. However, these effects were similar for all
Bokashi treatments, even if EM Bokashi increased the K con-
centrations in banana leaves significantly compared to Boka-
shi produced with EMst and the control. Bokashi produced
with only molasses and EM Bokashi decreased the number
of root nematodes under greenhouse conditions compared to
the control. Overall, the results confirmed the expected influ-
ence of composting on the degradation of organic material
and the effect of compost application on plant growth. Hower,
under the conditions of this study, EM showed no special
effects in this, except for increasing the K concentrations in
the leaves of young banana plants.
Key words: bokashi / compost / ergosterol / organic fertilizer /
microbial biomass
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1436-8730/07/0510-649
J. Plant Nutr. Soil Sci. 2007, 170, 649–656 DOI: 10.1002/jpln.200700002 649
1 Introduction
“Effective microorganisms” (EM) refer to an undisclosed mix-
ture of naturally occurring microorganisms that supposedly
have beneficial properties in a wide range of applications
(Higa, 2002). It was developed in the early 1980s by Tiruo
Higa, Professor at the University of Ryukyu, Okinawa, Japan,
and is sold as a commercial product. The exact composition
of EM has never been disclosed by the manufacturer. Work
of Kyan et al. (1999) showed that it predominantly contained
populations of lactic acid bacteria, photosynthetic bacteria,
yeasts, and Actinomycetes. According to Daly and Steward
(1999), 1 mL of the EM concentrate contains a minimum of
105viable organisms of the species Streptomyces albus,Pro-
pionibacterium freudenreichil,Streptococcus lactis,Aspergillus
oryzae,Mucor hiemalis,Saccharomyces cerevisiae,andCan-
dida utilis, in addition to an unspecified number of Lactobacillus
sp., Rhodopseudomonas sp., and Streptomyces griseus. Appli-
cation of EM supposedly leads to increases inthe microbial bio-
diversity of soils which enhances their quality and the growth,
yield, a nd quality of crop s (Higa and Parr, 1994).
Bokashi is the Japanese term for “fermented” organic matter
and is equivalent to compost used in traditional organic farm-
ing which is mostly prepared with the addition of EM. It can
be prepared under complete anaerobic or aerobic conditions.
The latter means that partial anaerobic conditions occur in
the middle of the compost pile while the outer layers remain
aerobic. The preparation of anaerobic Bokashi is made in
closed vessels while the preparation of aerobic Bokashi is
similar to traditional composting with additional usage of a
cover such as a jute bag, straw mat, or similar material (Kyan
et al., 1999). Enhanced decomposition of plant material after
addition of EM during Bokashi production has been proposed
as an innovative approach that allows the odorless break-
down of banana residues in as little as 3 weeks and the facili-
tation of a rapid recycling of plant nutrients (Shintani and
Tabora, 2000). In addition to the undisclosed composition of
EM ingredients, many of the described effects (Higa and
Parr, 1994; Wood et al., 1998; Xu et al., 2000) are not conclu-
sively proven (often due to insufficient control treatments) nor
are the experimental conditions well documented. Very little
is known about the underlying microbial processes such as
time courses of the C and N turnover and of bacterial and fun-
gal populations during the decomposition process. The
potential causes of the reported large yield increases,
* Correspondence: Prof. Dr. Andreas Buerkert;
e-mail: buerkert@uni-kassel.de
improvements of banana health, and increases in secondary
root growth of adult banana plants following the application of
EM Bokashi in plantations (Tabora et al., 2000) are not
known. Nevertheless, EM is used widely in agriculture pro-
duction in Asian countries and for Bokashi production on
banana farms in Cost Rica.
Therefore, the objective of this study was to fill the existing
gaps of knowledge about the effectiveness of EM Bokashi on
the compost quality and plant growth of bananas. The focus
of this study was based on the hypothesis that application of
EM increases mineralization processes of Bokashi through
an enhanced microbial colonization and activity. It was further
hypothesized that EM Bokashi increases growth of young
banana plants. To test these hypotheses, Bokashi of banana
residues was produced with daily applications of EM and
compared to three other non-EM treatments (water,
molasses, and sterilized EM) which allowed the differences
between the effects of added living organisms and pure sub-
strate effects to be distinguished. The produced Bokashi var-
iants were then applied to banana plants in a pot experiment.
2 Materials and methods
2.1 Study site
All experiments were conducted at EARTH University
(Escuela de Agricultura de la Región Trópico Húmedo) in
Costa Rica which is located between 83° and 84° W and at
10° N, 50 m asl. At this site, air temperatures range from
20°C to 30°C, with a mean relative humidity of 80% and
annual rainfall of approximately 3200 mm (EARTH, 2003).
2.2 Compost production
For the so called “activation” of EM following the product
description, 2 L of EM-stock solution were diluted in a closed
plastic barrel with 2 L of molasses and 60 L of tap water
(1:1:30). After 7 d, this mixture was applied to compost piles
(treatment EM). This solution contained 13.1 g L–1 dissolved
organic carbon (Corg), 244 mg L–1 nitrogen (N), 7.0 mg L–1
phosphorus (P), and 1.3 g L–1 potassium (K; Tab. 1).
Besides this EM treatment, three control treatments were
included in this study: A fixed amount of the diluted EM solu-
tion was heat-sterilized for 30 min at 121°C in an autoclave
(treatment EMst). Nutrient concentrations in EMst, after 7 d
of fermentation followed by sterilization, were higher than in
EM and pure molasses (Tab. 1). The latter was prepared
under the same conditions as EM but without effective micro-
organisms at a molasses-to-water ratio of 1:30 (treatment M).
This solution contained more dissolved organic carbon and P
than EM (Tab. 1). The fourth treatment consisted of pure tap
water (treatment W). These four Bokashi variants, with five
replications each, were produced by daily, early morning,
applications of 31mL of each treatment solution using a
spray bottle to achieve uniform surface coverage of the com-
post piles. All twenty Bokashi heaps, originally consisting of
120 kg of fresh banana material (chopped fruits and stalks),
were covered with approximately 10 kg sawdust and
arranged in a randomized block design on a concrete plat-
form protected by a roof.
Temperatures in the middle of the compost heaps were mea-
sured at 7:00 a.m., 11:30 a.m., and 5:00 p.m. every day with a
digital thermometer inserted to 20cm depth. On eight occa-
sions, the total weight of each heap was determined with a ten-
sion spring balance. For this, each heap was scooped in a big
plastic barrel and subsequently turned. Representative sam-
ples, consisting of mixed equal amounts of the outer and inner
compost layers, were taken on these eight occasions for analy-
sis of their nutrient and ergosterol concentrations. Microbial bio-
mass carbon (Cmic) and microbial biomass nitrogen (Nmic)as
well as ATP concentrations were only measured at the end of
the experiment to determine differences in microbial coloniza-
tion and activity at the end of the composting period.
2.3 Pot experiment
Three-week-old banana plants of Musa acuminata. (AAA) cv.
“Giant Cavendish”/“William” were planted in 22 L plastic pots
filled with a mixture of 16 L of a typical planting soil (containing
2.49% Corg) and 6 L of the four Bokashi variants (EM = produced
with activated EM; EMst = produced with sterilized activated
EM; M = produced with molasses; W = produced with water).
Each treatment, replicated six times, comprised one plant per
pot. The plants were placed in arandomized block design in the
greenhouse and rearranged every 4 d within blocks.
Over a period of 3 months, shoot growth (height and di-
ameter) was measured at weekly intervals with a measuring
tape. Finally, total plants were harvested, and shoot and root
fresh weight was determined. For this, roots were separated
from the adhering soil by carefully washing with tap water
over a sieve to prevent losses of fine roots. After cleaning
roots and leaves with demineralized water, they were dried at
65°C to weight constancy, ground, and analyzed for their total
N, P, and K contents.
2.4 Analytical measurements
2.4.1 Chemical and biological analyses of the compost
and plant samples
The ash content in the compost materials was determined
gravimetrically after heating at 550°C for 12 h in a muffle
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.plant-soil.com
Table 1: Total dissolved organic carbon (Corg), total nitrogen (N),
phosphorus (P), and potassium (K) concentrations in the solution of
effective microorganisms activated during 7 d of fermentation with
molasses and water at a ratio 1:1:30 (EM), in the sterilized solution of
activated EM (EMst) and in molasses (M), produced like EM but with-
out the addition of effective microorganisms.
Treatment
Corg
(g L–1)
N
(mg L–1)
P
(mg L–1)
K
(g L–1)
EM 13.1 244 7.0 1.3
EMst 14.1 272 9.3 1.5
M 13.7 246 7.8 1.3
650 Formowitz, Elango, Okumoto, Müller, Buerkert J. Plant Nutr. Soil Sci. 2007, 170, 649–656
oven. Organic-matter content was calculated as the differ-
ence between dry matter (24 h at 105°C) and ash content.
Total N was analyzed using an FP-328 N-analyser (LECO, St
Joseph, Mi, USA). For P and K analysis, dry matter was
determined after drying at 105°C for 24 h. Following combus-
tion at 550°C in a muffle oven, the ash was dissolved in
20 mL HCl (32%) and filled up to 100 mL after 12 h in the dark
with bi-distilled water. Subsequently, total P was measured
by spectrophotometry (U-2000 spectrophotometer, Hitachi,
Tokyo, Japan) using the vanadate-molybdate-method
(Gericke and Kurmies, 1952). Potassium was measured
using a flame photometer (Laboratory Instrument 543, Mas-
sachusetts, Lexington, USA).
For nematode counts, aliquots of 250 g root samples from
the banana plants of the greenhouse trial were washed and
then crushed for 10 s in a mixer filled with water. After sieving
the roots at 425 lm, 75 lm, and 45 lm, roots were filtered for
more then 24 h over a funnel (Bearmann funnel method;
Decker, 1969). Subsequently, an aliquot of 5mL was taken
and subsamples analyzed using a microscope under which
the nematodes were counted.
2.4.2 Microbial analyses of the compost samples
Compost samples were frozen and stored at –18°C from
sampling until analysis. Ergosterol as an indicator of fungal
biomass in the soils was determined according to Djajakirana
et al. (1996). Two grams of wet compost were extracted with
100 mL of ethanol by 30min oscillation shaking at 250
rev min–1 and then filtered (Whatman GF/A, UK). The extract
was dried at 40°C in a rotary evaporator and subsequently
dissolved in 9 mL of methanol (3 × 3 mL). After filtration
(0.45 lm cellulose-acetate filter, Sartorius AG, Göttingen,
Germany), ergosterol concentrations were measured by
reversed-phase HPLC analysis at 26°C.
Microbial biomass C and N were estimated using the chloro-
form-fumigation-extraction method (CFE; Brookes et al.,
1985; Vance et al., 1987) modified for compost samples
(Joergensen et al., 1996). After pre-extracting 25 g of
unsieved wet compost with 200 mL of 0.05 M K2SO4on a hor-
izontal shaker (200 rev min–1; 30 min) and manual removal of
larvae, the samples were filtered (Schleicher & Schuell
595½, Dassel, Germany). A subsample of 10 g of remaining
compost material was immediately fumigated for 24 h at
25°C with ethanol-free CHCl3. Fumigated and nonfumigated
samples were extracted with 100 mL 0.5 M K2SO4, filtered as
before and Cmic and Nmic calculated (Wu et al. 1990; Brookes
et al., 1985; Joergensen and Mueller, 1996).
The detection of adenine nucleotides (ATP, ADP, and AMP)
was carried out according to Dyckmans and Raubuch (1997)
using 4 g of wet compost. The adenylate energy charge
(AEC) was calculated as
AEC = (0.5 ADP + ATP) / (AMP + ADP + ATP).
2.5 Statistical analysis
All results were tested for normal distribution of residuals
using the Kolmogorov-Smirnov test. Compost variants were
compared using a GLM-repeated-measures ANOVA with the
treatments as between-subject factors. Variables were
regarded as innersubject factors, and means were separated
using Tukey’s honestly significant difference (HSD0.05). Tem-
perature measurements of the composts, plant growth (both
with time as a covariant), and plant samples were compared
using the GLM-univariate ANOVA, and means were sep-
arated using Tukey’s HSD0.05. All statistical analyses were
performed with SPSS 11.5 (Backhaus et al., 2003).
3 Results
3.1 Bokashi characteristics
All compost variants reached temperatures between 47°C
and 48°C after one day and up to 50°C on the second day of
composting. After the second turning of the heaps (8th day of
composting), temperatures reached 34°C–41°C. Following a
second thermophilic phase from the 15th to the 28th day, the
cooling or maturing phase started, and the temperatures
equaled ambient temperatures with no further temperature
changes. After 12 d, compost temperatures were significantly
different between EMst being the warmest and EM being the
coolest variant. The periodically measured sudden tempera-
ture drops coincided with the turning of the compost piles
(Fig. 1).
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.plant-soil.com
0 5 10 15 20 25 30 35
Average temperatures [ºC]
0
30
35
40
45
50
55 EM EMst M W
Days of composting
Figure 1: Average temperatures of Bokashi (compost)
produced with daily application of the four variants EM =
effective microorganisms, EMst = sterilized EM, M =
molasses, and W = water. Temperature measurements
were conducted at 7 a.m., 11:30 a.m., 5 p.m. each day.
Vertical bars represent ± one standard error of the mean.
Vertical arrows indicate the eight occasions when samples
were taken and the compost heaps turned.
J. Plant Nutr. Soil Sci. 2007, 170, 649–656 The role of “effective microorganisms” in composting 651
During 35 d of composting, all variants lost between 76.5%
and 78.3% of their initial fresh weight. The dry-matter content
of all Bokashi treatments decreased from 32 kg to approxi-
mately 11 kg during the first 15 d of composting. Afterwards,
dry matter slowly declined to a final value of approximately
7 kg (data not shown). The Corg concentrations were similar
for all treatments ranging from 44% to 50% during compost-
ing, until they dropped to 24.6% and 28.5% in Bokashi pro-
duced with molasses and with EM during the last week
(Tab. 2). However, statistical analysis indicated that EM Bokashi
contained significantly lower Corg concentrations than Boka-
shi produced with EMst and water. In EM Bokashi, the N con-
centration increased from 0.89% to finally 1.53%, whereas it
increased to 1.6%–1.7% in the other treatments. The C : N
ratios dropped from initially 53.0 to 29.0 and 25.6 for Bokashi
produced with water and with sterilized EM, and to 15.3 and
18.6 for Bokashi produced with molasses and EM, respec-
tively. Phosphorus increased only slightly in all treatments,
with initial concentrations of 0.21% and final concentrations
of 0.25%–0.27%, whereas K concentrations increased from
initially 4.1% to finally 6.4% to 6.6% (Table 2). No significant
differences were found for N, P, and K concentrations.
After 15 d of composting, the ergosterol concentration
reached its peak in all treatments with the highest concentra-
tions in Bokashi produced with water (105.3 lgg
–1) and the
lowest in EM (68.0 lgg
–1; Fig. 2). At the end of the compost-
ing process, this had changed, and the highest ergosterol
concentration was found in EM Bokashi (77.0 lgg
–1), while
the other variants had dropped to 29.6 lg–1 (EMst),
31.2 lgg
–1 (W), and 36.6 lgg
–1 (M). Microbial-biomass
C and Nmic were both lowest in EM (Cmic = 3121 lgg
–1;N
mic =
449 lgg
–1), while Cmic was highest in Bokashi produced with
molasses (3892 lgg
–1) and Nmic was highest in EMst
(615 lgg
–1; Tab. 3). However, these differences were not
statistically significant. The Cmic :N
mic ratio ranged from
5.9 to 7.7, with its lowest ratio in Bokashi produced with EMst
and the highest in Bokashi produced with molasses. The
ergosterol–to–biomass C ratio was highest in EM followed by
molasses (Tab. 3). Adenylate (ATP, ADP, and AMP) concen-
trations were lowest in EMst and highest in the molasses
treatment. The adenylate energy charge (AEC) was very
similar in all treatments (Tab. 3).
3.2 Pot experiment
Banana-shoot growth was similar in all treatments during the
first 3 weeks of the experiment. After 4 weeks, the growth
rates of all Bokashi treatments started to be higher than that
of the unamended control and between the 11th week and
harvest time, this difference was significant (data not shown).
Young banana plants grown in the control soil produced sig-
nificantly less shoot biomass (leaves and pseudo stem) than
plants grown with Bokashi, but significantly more root mass
(Tab. 4). Leaf analyses showed significantly higher P concen-
trations in banana grown in the Bokashi treatments than in
control plants, whereas no significant differences were found
for N concentrations (Tab. 4). Potassium concentrations were
significantly higher in banana leaves grown in the EM-Boka-
shi treatment compared to those of EMst Bokashi and of the
control (Tab. 4).
Root-borne nematodes were most frequent in the control
treatment, however, only differences between control on the
one hand and EM and M on the other hand were significant
(Fig. 3).
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.plant-soil.com
Table 2: Nutrient concentrations, expressed on a dry-matter basis, of
the initial banana material and the four different Bokashi variants at
the end of composting (EM = produced with effective microorgan-
isms; EMst = produced with sterilized EM; M = produced with
molasses; W = produced with water). Means with different letters in
each column are significantly different at p<0.05 (Tukey’s HSD).
Corg NPK C:N
Treatment (%)
Initial material 46.9 0.89 0.21 4.1 53.0
EM 28.5a1.53 0.25 6.6 18.6
EMst 43.6b1.71 0.25 6.5 25.6
M 24.6a1.62 0.25 6.4 15.3
W 48.6b1.67 0.27 6.6 29.0
Days of composting
0 5 10 15 20 25 30 35
Ergosterol concentrations [µg (g soil)-1]
0
10
20
30
40
50
60
70
80
90
100
110
120
130 EM EMst M W
Figure 2: Ergosterol concentrations during the
composting process of the following four Bokashi
(compost) variants: EM = produced with effective
microorganisms, EMst = produced with sterilized
EM, M = produced with molasses, and W =
produced with water. Vertical bars represent ±
one standard error of the mean.
652 Formowitz, Elango, Okumoto, Müller, Buerkert J. Plant Nutr. Soil Sci. 2007, 170, 649–656
4 Discussion
The measured temperatures in the compost piles were typi-
cal for the often reported mesophilic, thermophilic, and cool-
ing phases in organic-matter decomposition (Chefetz et al.,
1996; Butler et al., 2001). However, the higher temperatures
in the Bokashi produced with sterilized EM might be
explained by the nutrient contents of the 31mL applied EM
solution (Tab. 1). At the first day of application, EMst con-
tained 30 mg higher organic C, 9 mg higher N, 0.07 mg high-
er P, and 5 mg higher K contents compared to the activated
EM. These additional nutrients can be used by the naturally
occurring microorganisms for their reproduction, increasing
their activity and may thus have led to the higher tempera-
tures observed.
The observed high losses of fresh weight and dry matter dur-
ing the first 15 d of composting might be the result of the high
liquid losses and respiration processes. Consequently, a
decrease in C concentration and an increase in the ash con-
centration over time would have been expected as reported
by Daly and Stewart (1999) and Chefetz et al. (1996). Their
results are in contrast to the more or less stable concentra-
tions of organic C measured during this study. Our constant
C concentrations should therefore be interpreted with cau-
tion. However, in composts or other organic fertilizers C : N
ratios > 20 will normally lead to a temporary immobilization of
N through microorganisms and thus cause N deficiency in
plants (Akhtar, 2000; Lloyed et al., 2002). Therefore, Bokashi
produced with molasses and with EM, having C: N ratios
<20, seems to be a better plant-growth promoter than a
Bokashi produced with water and sterilized EM, both of which
had C : N ratios > 20.
Except for K, the nutrient concentrations of all Bokashi var-
iants were in the range of those measured in aerobically com-
posted or fermented composts made of different organic
materials (Bruns, 1996; Körner and Ritzkowski, 1999). When
compared to the values reported by Körner and Ritzkowski
(1999), the up to 22-fold higher K concentrations measured in
our Bokashi variants were probably due to the high K concen-
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.plant-soil.com
Table 3: Concentrations of microbial biomass C (Cmic), microbial biomass N (Nmic), ergosterol, adenylates (ATP, ADP, AMP), and AEC as well
as the ratios of Cmic :N
mic and ergosterol to Cmic at the last day of composting (36th day; EM = produced with effective microorganisms; EMst =
produced with sterilized EM; M = produced with molasses; W = produced with water). Means with different letters in each column are signifi-
cantly different at p<0.05 (Tukey’s HSD).
Cmic Nmic Ergosterol Cmic :N
mic Ergosterol (%) to
Cmic (%)
AMP ADP ATP AEC
Treatment (lgg
–1) (nmol g–1)
EM 3121 449 77.0a6.7 2.47a8.5 5.4 15.7 0.62
EMst 3636 615 29.6b5.9 0.81b8.0 3.7 13.6 0.61
M 3892 505 36.6b7.7 0.94b9.6 6.2 19.3 0.64
W 3846 576 31.2b7.0 0.81b8.5 4.9 15.1 0.62
Table 4: Dry matter (DM; 105°C) of shoots and roots and nutrient
concentrations (all three expressed on a dry-matter basis) of young
banana plants grown over 3 months under greenhouse conditions
with application of four Bokashi variants (EM = produced with effec-
tive microorganisms; EMst = produced with sterilized EM; M = pro-
duced with molasses; W = produced with water) compared to a con-
trol without Bokashi application. Means with different letters in each
column are significantly different at the level p<0.05 (Tukey’s HSD).
Leaf nutrient concentrations
Shoot DM Root DM N P K
Treatment (g) (mg g–1)
EM 1142b304a25.9 1.51b48.7c
EMst 1250b326a26.2 1.56b42.7b
M 1246b363a26.1 1.53b45.5bc
W 1182b321a26.0 1.56b43.5bc
Control 772a430b25.8 1.22a19.6a
Treatments
EM EMst M W Contr
Root born nematodes (100 g)-1
0
100
200
300
400 aabaabb
Figure 3: Nematode counts from root samples grown in the following
Bokashi (compost) treatments: EM = produced with effective
microorganisms, EMst = produced with sterilized EM, M = produced
with molasses, W = produced with water, and Contr = control without
Bokashi application. The vertical box plots show the median and the
25% and 75% percentiles as vertical boxes and 10% and 90%
percentiles as error bars. Minimum and maximum values are
indicated by black dots. The means with different letters are
significantly different at p<0.05 (Tukey’s HSD).
J. Plant Nutr. Soil Sci. 2007, 170, 649–656 The role of “effective microorganisms” in composting 653
trations in the raw banana material used for compost produc-
tion (Ultra Jr. et al., 2005). No significant effect of living EM
on mineralization processes was found in this study. This
might be due to the relatively small amount of EM applied
that was unlikely to dominate the microbial-community struc-
ture in the compost piles. In contrast, Schloss et al. (2003)
observed effects of EM on mineralization processes in the
treatment of waste water. Similarly, VanderGheynst and Lei
(2003) reported an influence of mixing and aeration on the
microbial-community structure of a rice straw– and dairy
manure–based compost. Assuming that the quality of Boka-
shi depends on the predomination of lactic acid fermentation,
which is decreased through increased aeration (Yamada and
Xu, 2000), it cannot be completely ruled out that the high fre-
quency of turning the compost piles practiced in our study
might have hampered the enhancement of mineralization
through the inoculum. The different material and slightly dif-
ferent composting technique, using saw dust as a cover for
the compost heaps instead of jute bags or straw mats, might
have led to differences in the chemical and microbiological
parameters characterizing the Bokashi produced in the pre-
sent study as compared to the one described by Yamada and
Xu (2000). On the other hand, our results are in line with
observations by Goto and Muramoto (1995). They compared
EM Bokashi from the EM company with ordinary Bokashi
made by farmers without using EM and did not find any differ-
ences between the two composts.
The measured ergosterol concentrations, which were highest
on the 15th day of composting in all variants, are in contrast
to findings reported in the literature, where the final commu-
nity structure of composts is dominated by fungi (Hellmann et
al., 1997; Mondini et al., 2002). Only EM Bokashi showed an
increase in fungi colonization on the final date of measure-
ment. Microbial C concentrations in Bokashi of about 3.1 to
3.9 mg g–1 on the final day of composting (36th day) are
almost 3-fold higher than those measured by Hellmann et al.
(1997) at the same time of composting, but comparable to
the Cmic concentrations of about 3 mg g–1 in a 9-week-old bio-
waste compost (Gattinger et al., 2004). This leads to the
assumption that the microbial population inside the Bokashi
piles was growing faster than in a traditional compost and
would thus confirm earlier findings that Bokashi composts are
ready for usage in a very short time (Kyan et al., 1999). In EM
Bokashi, in contrast, the significant lower temperatures with
significantly higher ergosterol and lower Cmic concentrations
at the end of composting compared to the other variants indi-
cates a change in the microbial-community structure towards
one dominated by fungi. This is in contrast to the first hypo-
thesis that mineralization was enhanced through increased
microbial colonization after the addition of EM.
Even though the AEC indicates dormant cells or microorgan-
isms in a stationary phase as found for in vitro cultures (AEC
<0.8; Brookes et al., 1983), the ATP concentrations inall Boka-
shi variants were almost 7 to 10 times higher (6.9–9.8 lgg
–1)
than those measured by Tseng et al. (1995) after 10 d of com-
posting. This indicates a higher microbial activity in the Boka-
shi variants in our study, which was probably due to the
7%–12% higher water content (67%–72%) inside the Bokashi
piles. The dormant cells (AEC <0.8) with large ATP concen-
trations might have been in a state of “metabolic alertness”
(De Nobili et al., 2000), which allows them to rapidly increase
their metabolism again if necessary, such as for capturing
nutrients suddenly available after turning the compost piles.
The usual increase in plant growth through compost applica-
tion was also found in this study for all Bokashi variants
compared to the control (De Brito Alvarez et al., 1995). Signif-
icantly more root mass was measured in the control without
any Bokashi application, probably due to the poor nutrient
status of the substrate that forced plants to develop more,
longer, and more active roots scavenging for nutrients (Xu,
2000). Hence, it could not be confirmed that Bokashi pro-
duced with EM promotes plant growth more than any of the
other Bokashi variants.
The ability of Bokashi to suppress nematodes might be partly
due to sterilizing compounds produced by lactic acid bacteria
that suppress harmful microorganisms, such as Fusarium
and nematodes (Kyan et al., 1999). Nematode-controlling
effects may also arise from the chemical composition of the
organic additives and the type of microorganisms that devel-
op during the decomposition process (Rodriguez-Kabana et
al., 1987). Organic amendments may also control nematodes
by direct toxicity (Akhtar and Alam, 1993). However, our data
did not confirm nematode-controlling effects through the addi-
tion of EM such as reported by Tabora et al. (2000). All Boka-
shi variants tended to reduce nematode infestation, even if
the observed effects were statistically significant only for EM
and M. This may lead to the assumption that the reduced
numbers of nematodes might be due to a more general Boka-
shi effect.
5 Conclusions
Specific effects of EM could only be observed for the temper-
ature development during composting and fungi population
indicated by ergosterol concentrations at the last sampling
date. This study did not provide evidence of EM-induced
enhanced mineralization nor increased colonization of com-
posts with added microorganisms. In the EM treatment rather
the opposite was observed with a fungi-dominated microbial
population at the end of composting. The results provide evi-
dence that the plant growth–promoting effect of added EM
Bokashi was based on the applied organic substrate rather
than the addition of EM. Thus, under the conditions of this
study (EM-application rates, frequency of compost-pile turn-
ing), effects of EM on the decomposition process and growth
of young banana plants were minor at most. Further research
under different composting conditions, with higher treatment
rates, an additional N source, and optimized banana-growing
conditions would be needed to verify our results.
Acknowledgments
We would like to thank EARTH University, Costa Rica and
the Dole Company for partial funding of this study. We are
also grateful to Gabriele Dormann and Herbert Arrieta Vargas
for their excellent technical assistance, to James Thompson
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.plant-soil.com
654 Formowitz, Elango, Okumoto, Müller, Buerkert J. Plant Nutr. Soil Sci. 2007, 170, 649–656
for correcting the English and to two reviewers for their help-
ful comments on earlier versions of this paper.
References
Akhtar, M. (2000): Approaches to biological control of nematode
pests by natural products and enemies. J. Crop Prod. 3, 367–395.
Akhtar, M., Alam, M. M. (1993): Utilization of waste materials in
nematode control: a review. Bioresour. Technol. 45, 1–7.
Backhaus, K., Erichson, B., Plinke, W., Weiber, R. (2003): Multi-
variate Analysemethoden – eine anwendungsorientierte
Einführung. Springer-Verlag, Berlin, Germany.
Brookes, P. C., Tate, K. R., Jenkinson, D. S. (1983): The adenylate
energy charge of the soil microbial biomass. Soil. Biol. Biochem.
15, 9–16.
Brookes, P. C., Landman, A., Pruden, G., Jenkinson, D. S. (1985):
Chloroform fumigation extraction and the release of soil nitrogen: a
rapid direct extraction method to measure microbial biomass in
soil. Soil Biol. Biochem. 17, 837–842.
Bruns, C. (1996): Suppressive Effekte von Komposten aus der
getrennten Sammlung organischer Abfälle und von Rindermist-
kompost gegenüber bodenbürtigen Schaderregern. PhD thesis,
Organic Agriculture (Ökologischer Landbau), University of Kassel.
Butler, T. A., Sikora, L. J., Steinhilber, P. M., Douglass, L. W. (2001):
Compost age and sample storage effects on maturity indicators of
biosolids compost. J. Environ. Quality 30, 2141–2148.
Chefetz, B., Hatcher, P. G., Hadar, Y., Chen, Y. (1996): Chemical and
biological characterization of organic matter during composting of
municipal solid waste. J. Environ. Quality 25, 776–785.
Daly, M. J., Stewart, D. P. C. (1999): Influence of “effective microor-
ganisms” (EM) on vegetable production and carbon mineralization
– a preliminary investigation. J. Sustain. Agric. 14, 15–25.
De Brito Alvarez, M. A., Gagné, S., Antoun, H. (1995): Effect of
compost on rhizosphere microflora of the tomato and on the inci-
dence of plant growth-promoting rhizobacteria. Appl. Environ.
Microbiol. 61, 194–199.
De Nobili, M., Contin, M., Mondini, C., Brookes, P. C. (2000): Soil
microbial biomass is triggered into activity by trace amounts of
substrate. Soil Biol. Biochem. 33, 1163–1170.
Decker, H. (1969): Phytonematologie. VEB Deutscher Land-
wirtschaftsverlag, Berlin, Germany.
Djajakirana, G., Joergensen, R. G., Meyer, B. (1996): Ergosterol and
microbial biomass relationship in soil. Biol. Fert. Soils 22,
299–304.
Dyckmans, J., Raubuch, M. (1997): A modification of a method to
determine adenosine nucleotides in forest organic layers and
mineral soils by ion-paired reversed-phase high-performance liquid
chromatography. J. Microbiol. Methods 30, 1–20.
EARTH (Escuela de Agricultura de la Región Tropical Húmeda)
(2003): www.earth.ac.cr. Date of access: May 13, 2003.
Gattinger, A., Bausenwein, U., Bruns, C. (2004): Microbial biomass
and activity in composts of different composition and age. J. Plant
Nutr. Soil Sci. 167, 556–561.
Gericke, S., Kurmies, B. (1952): Die kolorimetrische Phosphorsäure-
bestimmung mit Ammonium-Vanadat-Molybdat und ihre
Anwendung in der Pflanzenanalyse. Z. Pflanzenernähr. Düng.
Bodenk. 59, 235–247.
Goto, I., Muramoto, J. (1995): Chemical properties of EM bokashi, in
Nishio, M.: Microbial fertilizer in Japan 11. Proceedings of the
Japanese Society of Soil Science and Plant Nutrition 41, p. 158.
Goto, I., Muramoto, J. (1995): Chemical properties of EM bokashi., in
Nishio, M. (1996): Microbial fertilizer in Japan, 11. Proceedings of
the Japanese Society of Soil Science and Plant Nutrition 41,
p. 158; www.eprida.com/hydro/ecoss/background/microbefert.pdf.
Date of access: November 15, 2006.
Hellmann, B., Zelles, L., Palojarvi, A., Bai, Q. Y. (1997): Emission of
climate-relevant trace gases and succession of microbial commu-
nities during open-window composting. Appl. Environ. Microbiol.
63, 1011–1018.
Higa, T. (2002): Eine Revolution zur Rettung der Erde: mit effektiven
Mikroorganismen (EM) die Probleme unserer Welt lösen. OLV
(Organischer Landbau Verlag), Kevelaer, Germany.
Higa, T., Parr, J. F. (1994): Beneficial and effective microorganisms
for a sustainable agriculture and environment. INFRC (Interna-
tional Nature Farming Research Center), Atami, Japan.
Joergensen, R. G., Mueller, T. (1996): The fumigation-extraction
method to estimate soil microbial biomass: calibration of the kEN
value. Soil Biol. Biochem. 28, 33–37.
Joergensen, R. G., Meyer, B., Roden, A., Wittke, B. (1996): Microbial
activity and biomass in mixture treatments of soil and biogenic
municipal refuse compost. Biol. Fertil. Soils 23, 43–49.
Körner, I., Ritzkowski, M. (1999): Charakteristik von Bioabfällen und
Komposten, in Körner, I., Ritzkowski, M., Stegmann, R.: Neue
Techniken zur Kompostierung, Verwertung auf landwirtschaftlichen
Flächen, Teilvorhaben 10.b Teil 1 und Teil 2, Band II. Umweltbun-
desamt Hamburg, Abfallwirtschaft und Altlastensanierung des
BMBF, pp. 73–95.
Kyan, T., Shintani, M., Kanda, S., Sakurai, M., Ohashi, H., Fujisawa,
A., Pongdit, S. (1999): Kyusei nature farming and the technology
of effective microorganisms, guidelines for practical use.
Sangakkara, R. (ed.), APNAN (Asia Pacific Natural Agriculture
Network, Bangkok, Thailand) and INFRC (International Nature
Farming Research Center), Atami, Japan.
Lloyed, J. E., Herms, D. A. H., Stinner, B. R., Hoitink, H. A. J. (2002):
Comparing composted yard trimmings and ground wood as
mulches. Bio Cycle 43, 52–56.
Mondini, C., Contin, M., Leita, L., De Nobili, M. (2002): Response of
microbial biomass to air-drying and rewetting in soils and compost.
Geoderma 105, 111–124.
Rodriguez-Kabana, R., Morgan-Jones, G., Chet, I. (1987): Biological
control of nematodes: soil amendments and microbial antagonists.
Plant Soil 100, 237–247.
Schloss, P. D., Hay, A. G., Wilson, D. B., Walker, L. P. (2003): Molec-
ular assessment of inoculum efficacy and process reproducibility in
composting using ARISA. Transactions of the ASAE 46, 919–927.
Shintani, M., Tabora, P. (2000): Organic fertilizer: managing banana
residues with effective microorganism (EM), in Alföldi, T., Lock-
eretz, W., Niggli, U.: IFOAM 2000 – The world grows organic.
Proceedings 13th International IFOAM Scientific Conference,
Basel, 28–31 August 2000. Vdf Hochschulverlag AG an der ETH
Zurich, Switzerland, p. 269.
Tabora, P., Shintani, M., Elango, F. (2000): Banana researches in
Costa Rica (Central America) with Effective Microorganisms.
www.emro.jp./english/library/abstracts/0006banana.html. Date of
access: June 25, 2003.
Tseng, D. Y., Chalmers, J. J., Tuovinen, O. H., Hoitink, H. A. J.
(1995): Characterization of a bench-scale system for studying the
biodegradation of organic solid wastes. Biotechnology 11,
443–451.
Ultra Jr., V. U., Mendoza, D. M., Briones, A. M. (2005): Chemical
changes under aerobic composting and nutrient supplying
potential of banana residue compost. Ren. Agric. Food Sci. 20,
113–125.
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.plant-soil.com
J. Plant Nutr. Soil Sci. 2007, 170, 649–656 The role of “effective microorganisms” in composting 655
Vance, E. D., Brookes, P. C., Jenkinson, D. S. (1987): An extraction
method for measuring soil microbial biomass C. Soil Biol. Biochem.
19, 703–707.
VanderGheynst, J. S., Lei, F. (2003): Microbial community structure
dynamics during aerated and mixed composting. Transactions of
the ASAE 46, 577–584.
Wood, M. T., Miles, R., Tabora, P. (1998): EM-fermented plant extract
and EM5 for controlling pickleworm (Ciaphania nitidalis) in organic
cucumber, in Foguelman, D., Lockeretz, W. (eds.): IFOAM 98 –
Organic agriculture the credible solution for the XXIst Century.
Proceedings of the 12th International IFOAM Scientific
Conference, Mar del Plata/ Argentinien, November 1998, IFOAM
c/o Ökozentrum Imsbach, Tholey-Theley, Germany, pp. 219–224.
Wu, J., Joergensen, R. G., Pommerening, B., Chaussod, R.,
Brookes, P. C. (1990): Measurement of microbial biomass C by
fumigation extraction – an automated procedure. Soil Biol.
Biochem. 22, 1167–1169.
Xu, H. L. (2000): Effects of a microbial inoculant and organic ferti-
lizers on the growth, photosynthesis and yield of sweet corn.
J. Crop Prod. 3, 183–214.
Xu, H. L., Wang, R., Mridha, M. (2000): Effects of organic fertilizers
and a microbial inoculant on leaf photosynthesis and fruit yield and
quality of tomato plants. J. Crop Prod. 3, 173–182.
Yamada, K., Xu, H. L. (2000): Properties and applications of an
organic fertilizer inoculated with effective microorganisms. J. Crop
Prod. 3, 255–268.
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.plant-soil.com
656 Formowitz, Elango, Okumoto, Müller, Buerkert J. Plant Nutr. Soil Sci. 2007, 170, 649–656
