Content uploaded by Nevena Čule
Author content
All content in this area was uploaded by Nevena Čule on Dec 20, 2016
Content may be subject to copyright.
© by PSP Volume 25 – No. 9/2016, pages 3728-3733 Fresenius Environmental Bulletin
3728
PHYTOREMEDIATION POTENTIAL OF CANNA INDICA L.
IN WATER CONTAMINATED WITH LEAD
Nevena Cule1*, Dragica Vilotic2, Marija Nesic2, Milorad Veselinovic1, Dragana Drazic1, Suzana Mitrovic1
1 Institute of Forestry, Kneza Višeslava 3, 11030 Belgrade, Serbia
2Faculty of Forestry, University of Belgrade, Kneza Višeslava 1, 11030 Belgrade, Serbia
ABSTRACT
Today there are many technologies for
wastewater treatment and rhizofiltration is one of
phytoremediation techniques that is very promising
for cleanup of large quantities of water with
medium or low concentrations of heavy metals. The
aim of this study was to investigate
phytoremediation potential of ornamental plant C.
indica in water contaminated with lead. The present
research demonstrated that dry weight of above-
ground and below-ground biomass was
significantly increased at the highest treatment
containing 41 mgPb/L. Lead accumulation in
below-ground biomass was up to 90- fold higher
than in above-ground biomass. The highest Pb
concentration was recorded in root (2480.07 mg/kg)
on the 21st sampling day in treatment with the most
Pb added. The highest bioconcentration factor
(81.16) was recorded in the nutrient solution with
the least Pb added. Translocation factor was not
significantly affected by lead concentration in
nutrient solution or exposure time and it was low
(0.01). Symptoms of lead phytotoxicity were not
observed on any plant in treatments and control.
The results of this research further support the idea
that terrestrial plants are more suitable for
rhizofiltration than aquatic plants and that C. indica
can be used in rhizofiltration systems or floating
islands for treatment of water polluted with lead.
KEYWORDS:
rhizofiltration, lead, Canna indica L., biomass,
bioconcentration factor, translocation factor
INTRODUCTION
Plants require a considerable number of metals
in very small quantities for their growth and
development. However, some of biogenic elements,
such as Cu, Se and Zn, are toxic in high
concentrations and may be found in some types of
wastewater. Other metals such as Cd, Hg, Pb, As,
Tl and U can also be detected in industrial and other
wastewater, but they have no biological value to
living organisms, and are extremely toxic in
relatively low concentrations [1,2]. These metals
have an adverse effect to the environment and
humans because they can easily travel through the
food chain and accumulate for many years in its
highest links [3].
Rhizofiltration is one of phytoremediation
techniques that uses plant roots for the absorption,
concentration and precipitation of metals from
water [4]. Terrestrial plants are considered more
suitable for rhizofiltration because they have longer,
stronger and usually fibrous roots with a large
surface area for metal sorption [5]. Canna indica L.
(Cannaceae) may be a good candidate for heavy
metal removal from polluted water because it has
several important features of plants suitable for
phytoremediation.
The aim of this study was to investigate
phytoremediation potential of C. indica. To achieve
this aim main objectives were to grow plants in
nutrient solution with three different concentrations
of lead, to determine dry biomass of vegetative
parts, to analyse plants for lead concentration, to
compare lead concentration in above-ground and
below-ground biomass, to quantify the potential of
plant to concentrate desired metal from the medium,
and to determine the ability of the plant to
translocate heavy metal from roots to above-ground
biomass.
MATERIALS AND METHODS
The plant material was established from
rhizomes (with the approximately same number of
buds and weight) of the species C. indica which
were planted in peat at the beginning of April.
Planted material was stored in the laboratory with a
glass roof till the beginning of the experiment (the
end of May). Light regime was consistent with
normal alternation of day and night.
Well-cultivated seedlings with an average
height of 50 cm were transferred to containers filled
with 3L of modified half strength Hoagland
solution. Nutrient solution contained (mM): 2
Ca(NO3)2 x 4H2O, 3 KNO3, 3 NH4NO3, 1 MgSO4 x
7H2O, 1 KH2PO4 and 0,2 FeEDTA, and (µM) 4,5
MnSO4, 23 H3BO3, 0,1 (NH4)6Mo7O24, 0,4 ZnSO4
and 0,2 CuSO4 [6]. Solutions were changed and
© by PSP Volume 25 – No. 9/2016, pages 3728-3733 Fresenius Environmental Bulletin
3729
supplemented with new ones after a week. Lead
was supplied as Pb(CH3COO)2 x 3H2O to the
medium at three different concentrations: 50 µM ,
100 µM and 200 µM containing 10 (treatment I),
21(treatment II) and 41 (treatment III) mgPb/L
solution, respectively. Plants grown in nutrient
solution without lead presented control.
Plant samples were collected on 9th and 21st
day of the experiment. Plants were removed from
nutrient solution and divided into roots, rhizomes,
stems and leaves. All vegetative parts were rinsed
three times in distilled water to remove surface
adsorbed lead. Plants were then transferred to paper
bags and dried at 80oC for 24 hours [7]. Plant
samples were milled to powder to pass 20-mesh
sieve. Representative samples were placed in plastic
containers and stored for chemical analysis.
The fresh and dry weights of vegetative parts
were measured on electronic balance in order to
determine above-ground and below-ground
biomass.
The heavy metals were extracted using a
microwave digestion method described by Senila et
al. [8]. Microwave digestion unit (CEM MDS 2000,
Berghof, Germany, Mod. Speedwave MWS3+) was
used for sample preparation. Samples of a dry well
homogenised plant material (250-300 mg) were
weighted in Teflon vessels and 5 ml of 69% HNO3
and 2 ml of 30% H2O2 were added. Closed vessels
were placed in a microwave digester. Cooled
samples were decanted in 25mL volumetric flack
and were made up to the graduated line with
distilled water [9]. Samples were then filtrated with
filter paper and stored in closed sterile containers in
refrigerator for heavy metal analysis.
Heavy metal content in plant tissue was
analysed using ICP-OES (Varian Vista-PRO, CCD
Simultaneous ICP-OES). All tissue heavy metal
concentrations are reported on a dry weight basis.
Bioconcentration factor (BCF) was used to
quantify the potential of C. indica to concentrate
desired metal from the medium. It represents the
index that indicates the possibility of the plant to
accumulate the metal of interest in relation to its
concentration in the nutrient solution [10].
Bioconcentration factor (BCF) was calculated as:
BCF
= Pb concentration in dry plant tissue (mg/kg) at harvest
initial Pb concentration in nutrient solution (mg/L)
In rhizofiltration BCF of below-ground
biomass is more relevant measurement of plants
phytoremediation potential then BCF of whole
plant or BCF of above-ground biomass [11]. For
that reason BCF was separately calculated for both
above-ground and below-ground biomass. The
higher the value of bioconcentration factor is the
plant is more suitable for phytoremediation of
targeted heavy metal [12].
Translocation factor (TF) was calculated to
determine the potential of C. indica for
phytoremediation. It represents the index that
indicates the ability of the plant to translocate heavy
metals from roots to above-ground biomass [13].
Translocation factor (TF) was calculated as:
TF
= Pb concentration in dry above ground biomass (mg/kg)
Pb concentration in dry below ground biomass (mg/kg)
Values of translocation factor less than 1
indicates that the heavy metal accumulates largely
in below-ground biomass and its translocation to
the above-ground biomass is poor [14].
Influences of various treatments on biomass
dry weight, lead concentration in biomass,
bioconcentration factor and translocation factor
were tested using one-way analysis of variance
(ANOVA) followed by Fisher's LSD test (P <
0.05). All statistical analyses were carried out using
Statgraphics Centurion XVI (Statpoint
Technologies, Inc., Warrenton, VA, USA).
RESULTS AND DISCUSSION
The results obtained from the analysis of
effects of treatments and sampling day on the dry
weight biomass of C. indica are presented in Fig.1.
The present study demonstrated that dry weight of
above-ground and below-ground biomass were not
significantly affected by lead concentration in
control and treatments I and II on both sampling
days. Significant increase of dry weight biomass
was determined only at the highest treatment
containing 41 mgPb/L.
FIGURE 1
Effects of treatments on above-ground and
below-ground biomass.
ab bba
bbb
a
bb
b
a
bbb
a
0
2
4
6
8
10
12
0 I II III
Biomass (g)
Treatmen Pb conc. (mg /L)
AGB 9
AGB 21
BGB 9
BGB 21
© by PSP Volume 25 – No. 9/2016, pages 3728-3733 Fresenius Environmental Bulletin
3730
Within each graph different letters indicate
significant differences between treatments on the
same sampling day. Error bars represent ± SE of the
mean of 6 replicates. (AGB9 and AGB21 - above-
ground biomass on 9th and 21st sampling day,
respectively; BGB9 and BGB21 - below-ground
biomass on 9th and 21st sampling day, respectively)
The reason for this is not clear but biomass
weight could be affected by changes in enzyme
activity, mineral nutrition, hormonal status or
membrane permeability [15]. Although, our results
differ from some published studies [16,17,18,19],
they are consistent with those of De Jesus and
Yllano [10] who showed significant increases in the
root and shoot biomass of Zea mays L. in the
highest Pb treatment where Pb was supplied as
Pb(NO3)2.
Canna indica L. regulated lead uptake so that
concentration in plant tissue reflected levels of lead
in growth media (Tab. 1 and Tab. 2). The content of
lead in plants also increased with exposure time.
This was in agreement with data obtained by Bose
et al. [20] and De Jesus and Yllano [10].
The highest Pb accumulation in above-ground
biomass was 26.39 mg/kg at highest Pb supply on
the 21st sampling day (Tab. 1). The leaf and steam
contribute to this value was almost equal with 12.16
mg/kg and 14.23 mg/kg, respectively (Tab. 2).
These results are consistent with those of
Subhashini and Swamy [21] for lead accumulation
in C. indica and Qian et al. [22] for several wetland
plants such as Cyperus alternifolius L., Marsilea
drummondii A.Braun, Myriophyllum
brasiliense Camb. The accumulation of lead in
above-ground biomass was much lower compared
to Z. mays and Brassica juncea but close to lead
accumulated by B. juncea cv. 184290 [23].
Lead accumulation in below-ground biomass
was up to 90- fold higher than in above-ground
biomass. The same trend of accumulation was
observed in Z. mays [10] and B. juncea [23]. The
highest Pb bioaccumulation in below-ground
biomass was 2807.85 mg/kg at the highest Pb
supply on the 21st sampling day (Tab. 1). The root
accumulated much more Pb then rhizome with
2480.07 mg/kg and 327.77 mg/kg, respectively
(Tab. 2). These results are in line with those
obtained for Z. mays [10]. However, C. indica in
this study accumulated more Pb compared to other
findings for C. indica [21], Z. mays [23] and
wetland plants such as Polygonum hydropiperoides
Michx., C. alternifolius, M. drummondii and M.
brasiliense [22]. Dushenkov et al. [5] reported that
B. juncea was among the best accumulators in their
study with 136000 mgPb/kg thus C. indica in this
study was far behind stated results for B. juncea.
Results (Tab. 3) showed that different
treatments had an effect on BCF of above- ground
and below-ground biomass on the 9th sampling day
(p<0.05). However, with the duration of the
experiment differences in BCF among treatments
were reduced both for the above-ground and for the
below-ground biomass (p>0.05).
TABLE 1
Mean Pb concentration in above-ground and below-ground biomass (mg/kg)
Treatment
Pb concentration (mg/kg)
Above-ground biomass
Below-ground biomass
9 day
21 day
9 day
21 day
I
4.32±0.560b
8.37±1.397b
278.61±32.876b
840.89±73.535b
II
13.79±2.454a
20.84±4.900a
1178.41±167.637a
1536.10±213.770b
III
15.87±2.348a
26.39±3.634a
1216.05±60.073a
2807.85±408.962a
Values with a different letters within column and on the same sampling day are significantly different at <0.05.
Values are means ± SE of six replicates. TABLE 2
Mean Pb concentration in different vegetative parts (mg/kg)
Sampling
day
Vegetative
part
Treatments
I
II
III
9
Leaf
2.68±0.441b
7.75±1.131b
11.41±2.462c
Steam
1.63±0.293b
6.04±1.499b
4.46±0.897c
Rhizome
39.58±6.626b
96.04±21.369b
144.92±21.770b
Root
239.03±28.365a
1082.37±149.926a
1071.12±57.566a
21
Leaf
3.97±0.230b
6.40±1.281b
12.16±0.926b
Steam
4.39±1.234b
14.43±3.773b
14.23±3.125b
Rhizome
75.55±3.979b
129.72±32.350b
327.77±99.875b
Root
765.34±72.447a
1406.38±188.990a
2480.07±374.995a
Values with a different letters within column and on the same sampling day are significantly different at <0.05.
Values are means ± SE of six replicates.
© by PSP Volume 25 – No. 9/2016, pages 3728-3733 Fresenius Environmental Bulletin
3731
TABLE 3
Effects of treatments on BCF of above-and below-ground biomass and TF
Treatment
BCF of above-ground biomass
BCF of below-ground biomass
TF
9 day
21 day
9 day
21 day
9 day
21 day
I
0.41±0.053b
0.80±0.134a
26.89±3.174b
81.16±7.099a
0.02±0.001a
0.01±0.001a
II
0.67±0.120a
1.02±0.239a
57.70±8.209a
75.22±10.468a
0.01±0.001a
0.01±0.001a
III
0.38±0.056b
0.63±0.087a
29.34±1.449b
67.75±9.868a
0.01±0.001a
0.01±0.002a
Values with a different letters within column and on the same sampling day are significantly different at <0.05.
Values are means ± SE of six replicates
Contrary to the highest value of Pb
accumulation in the treatment with the highest Pb
supply, the highest BCF of below-ground biomass
(81.16) was recorded in the nutrient solution with
the least Pb added. In this study C. indica had
greater BCF compared to same plant in the
experiment carried by Subhashini and Swamy [21]
where BCF of whole plant did not reach more than
3.64. However, this maximum BCF value was
much lower than root BCF of Z. mays, B. juncea
and B. juncea cv. 184290 [23].
The above-ground biomass BCF was much
lower compared to below-ground biomass. This
further support finding of this study that much more
lead was accumulated from nutrient solution and
stored in roots than translocated to rhizome, steam
and leaves. The highest above-ground biomass BCF
was 1.02 in treatment II with medium Pb supply. As
with below-ground biomass BCF this value was
being far less than shoot BCF of the Z. mays, B.
juncea and B. juncea cv. [23].
The current study found that TF was not
significantly affected by lead concentration in
nutrient solution or exposure time (Tab. 3).
Translocation of lead was restricted from roots to
above-ground biomass thus TF was very low with a
value of 0.01 in all treatments and sampling days
except in treatment with the lowest Pb supply on
the 9th sampling day where it reached value of 0.02.
Several reports have also shown low TF for C.
indica [20], Brassica napus L. [13] and Z. mays, B.
juncea and B. juncea cv. 184290 [23]. It seems
possible that limited transport of lead from root to
other vegetative parts is due to the root endodermis
which acts as a barrier permitting lead entrance into
the central cylinder [15]. Seregin et al. [24] point
out that at the lethal concentrations of Pb the
plasmalemma is damaged, its barrier function is
broken and thus the greater amount of heavy metals
moves in to the symplast. It appears that Pb was
not supplied in lethal doses in present experiment so
low concentrations of Pb were found in above-
ground biomass compared to below-ground
biomass and thus TF was very low.
Based on the results of this study it could be
argued that C. indica may be classified as an
indicator plant in accordance to its ability to absorb,
accumulate and tolerate lead in its tissue. The plant
had a lower heavy metal accumulation as compared
to hyperaccumulators but it can produce at least ten
times greater biomass and thus the amount of the
accumulated heavy metals from contaminated
media is much higher. Furthermore, as a terrestrial
plant C. indica is also more suitable for
rhizofiltration than aquatic plants. The limited
rhizofiltration potential of aquatic macrophytes is
attributed to their relatively small root and its slow
growth rate in addition to the high water content in
their tissue which complicates their drying [5].
Results of our previous studies [25] showed that C.
indica can produced a significant amount of
biomass both of the above-ground and below-
ground biomass after a month and a half of growing
in water without the addition of any nutrients.
Furthermore, the plants were able to develop very
dense, strong fibrous roots with a large area for the
sorption of heavy metals. These findings further
suggest that C. indica may be tolerant to poor
environmental conditions.
The visual non-specific symptoms of lead
phytotoxicity [26] were not observed on any plant
in treatments and control. This finding was
unexpected and it may be the first step in
suggesting the strong possibility that C. indica may
also be tolerant to high levels of lead. The results of
this study corroborate the findings of Zurayk et al.
[11] who suggested that ability to tolerate high
levels of targeted heavy metal may be associated
with an increase of phytoaccumulation, with the
increase of heavy metal supply and restricted
translocation of heavy metal from root to shoot.
CONCLUSION
The findings of this study strongly suggest that
it is possible that C. indica is a very good candidate
for rhizofiltration of water contaminated with Pb.
The plant was tolerant to high Pb concentration in
nutrient solution thus the plant growth was not
impaired. Based on this fact and previous research
[25] it can be concluded that C. indica can grow
fast, rapidly produce large biomass and a dense root
system in unpolluted as well as in contaminated
water. Although, accumulation of Pb was not high
compared to B. juncea, C. indica performance was
generally better than commonly used plant species
for phytoremediation of Pb [1,10,14,16,22,23]. C.
© by PSP Volume 25 – No. 9/2016, pages 3728-3733 Fresenius Environmental Bulletin
3732
indica was able to limit root to shoot translocation
of Pb thus prevent the entry of lead into the food
chain, hinder its bio-magnification in the
environment and reduce the amount of secondary
waste at the end of rhizofiltration process. C. indica
is widespread ornamental species that is well
adapted to various climatic conditions. And finally,
the establishment and cultivation of the C. indica
seedlings were very simple. The results of this
research further support the idea that terrestrial
plants are more suitable for rhizofiltration than
aquatic plants and that C. indica can be used in
rhizofiltration systems or floating islands for
treatment of water polluted with lead.
ACKNOWLEDGMENTS
This paper was realized as a part of the
doctoral dissertation by Nevena Cule
“Phytoremediation of polluted water by plant
Canna indica L. and selected decorative
macrophytes” at University of Belgrade.
REFERENCES
[1] Kamal, M., Ghalya, A.E., Mahmouda, N.,
Cote, R. (2004) Phytoaccumulation of heavy
metals by aquatic plants. Environment
international, 29 (8), 1029-1039.
[2] Sekulić, P., Kastori, R., Hadžić, V. (2003)
Zaštita zemljišta od degradacije. Novi Sad,
Naučni institut za ratarstvo i povrtarstvo. (in
Serbian)
[3] Gutiérrez-Ginés, M.J., Pastor, J. and Hernández
A.J. (2010) Effect of heavy metals from mine
soils on Avena sativa l. and education
strategies. Fresenius Environmental Bulletin
19, 9b, 2083-2086.
[4] Zhu, Y.L., Zayed, A.M., Quian, J.H., De
Souza, M. and Terry, N. (1999)
Phytoaccumulation of trace elements by
wetland plants: II. Water hyacinth. Journal of
Environmental Quality, 28 (1), 339-344.
[5] Dushenkov, S., Kumar, P.B.A. N., Motto and
H., Raskin, I. (1995) Rhizofiltration: the Use of
Plants to Remove Heavy Metals From Aqueous
Streams. Environmental Science Technology
29(5), 1239-1245.
[6] Zayed, A., Lytle, C.M., Qian, J.-H., Terry, N.
(1998) Chromium accumulation, translocation
and chemical speciation in vegetable crops.
Planta, 206(2), 293-299.
[7] Campbell, C.R. and Plank, C.O. (1998)
Preparation of plant tissue for laboratory
analysis. In: Kalra, Y.P. (ed.) Handbook of
reference methods for plant analysis. Boca
Raton: CRC Press, Taylor & Francis Group,
37-49.
[8] Senila, M., Senila, L., Roman, C. (2011)
Evaluation of performance parameters for trace
elements analysis in perennial plants using
ICP-OES technique. Journal of Plant
Development, 18, 87-93.
[9] Aksoy, A. and Sahin, U. (1999) Elaeagnus
angustifolia L. as a Biomonitor of Heavy Metal
Pollution. Tr. J. of Botany 23(2), 83-87.
[10] De Jesus, B.R. and Yllano, O.B. (2005)
Bioaccumulation and Bioconcentration of Pb in
the Tissues of Zea mays L. Philippine Journal
of Science 134 (1), 21-29.
[11] Zurayk, R., Sukkariyah, B., Baalbaki, R.,
Ghanem, D.A. (2002) Ni Phytoaccumulation
in Mentha aquatica L. and Mentha sylvestris L.
Water, Air & Soil Pollution, 139(1),355-364.
[12] Blaylock, M.J., Salt, D.E., Dushenkov, S.,
Zakharova, O., Gussman, C., Kapulnik, Y.,
Ensley, B.D. and Raskin. I. (1997) Enhanced
accumulation of Pb in Indian mustard by soil-
applied chelating agents. Environmental
Science and Technology, 31 (3), 860- 865.
[13] Marchiol, L., Assolari, S., Sacco, P., Zerbi, G.
(2004) Phytoextraction of heavy metals by
canola (Brassica napus) and radish (Raphanus
sativus) grown on multicontaminated soil.
Environmental Pollution 132(1), 21-27.
[14] Mellem, J.J., Baijnath, H., Bharti Odhav, B.
(2009) Translocation and accumulation of Cr,
Hg, As, Pb, Cu and Ni by Amaranthus dubius
(Amaranthaceae) from contaminated sites.
Journal of environmental science and health.
Part A, Toxic/hazardous substances &
environmental engineering, 44 (6), 568–575.
[15] Sharma, P. and Dubey, R,S. (2005) Lead
toxicity in plants. Brazilian Journal of Plant
Physiology, 17(1), 35-52.
[16] Wang, Q., Cui, Y., Dong, Y. (2002)
Phytoremediation of Polluted Waters Potentials
and Prospects of Wetland Plants. Acta
Biotechnologica, 22 (1-2), 199-208.
[17] Mishra, A. and Choudhuri, M,A. (1998)
Amelioration of lead and mercury effects on
germination and rice seedling growth by
antioxidants. Biologia Plantarum, 41(3), 469-
473.
[18] Breckle, C.W. (1991) Growth under heavy
metals. In: Waisel, Y., Eshel, A., Kafkafi, U.
(eds.) Plant roots: the hidden half. New York,
NY: Marcel Dekker, 351-373.
[19] Obroucheva, N.V., Bystrova, E.I., Ivanov,
V.B., Antipova, O.V., Seregin, I.V. (1998)
Root growth responses to lead in young maize
seedlings. Plant and Soil, 200(1), 55-61.
[20] Bose, S., Jain, А., Rai, V., Ramanathan, A.L.
(2008) Chemical fractionation and
translocation of heavy metals in Canna indica
L. grown on industrial waste amended soil.
Journal of Hazardous Materials 160(1), 187-
193.
© by PSP Volume 25 – No. 9/2016, pages 3728-3733 Fresenius Environmental Bulletin
3733
[21] Subhashini, V. and Swamy, A.V.V.S. (2014)
Phytoremediation of Metal (Pb, Ni, Zn, Cd
And Cr) Contaminated Soils Using Canna
Indica. Current World Environment, 9(3), 780-
784.
[22] Qian, J.H., Zayed, A., Zhu, Y.L., Yu, M.
and Terry, N. (1999) Phytoaccumulation of
Trace Elements by Wetland Plants: III. Uptake
and Accumulation of Ten Trace Elements by
Twelve Plant Species. Journal of
Environmental Quality, 28(5), 1448-1455.
[23] Huang, J.W. and Cunningham, S.D. (1996)
Lead phytoextraction: Species variation in lead
uptake and translocation. New Phytologist,
134(1), 75-84.
[24] Seregin, I.V., Shpigun, L.K., Ivanov,
V.B. (2004) Distribution and Toxic Effects of
Cadmium and Lead on Maize Roots. The
Russian Journal of Plant Physiology, 51(4),
525-533.
[25] Cule, N., Jovanovic, Lj., Drazic, D.,
Veselinovic, M., Mitrovic, S., Nesic, M. (2011)
Indian Shoot (Canna indica L.) In
Phytoremediation Of Water Contaminated
With Heavy Metals. Sustainable forestry, 63-
64, 51-64.
[26] Burton, K.W., Morgan, E., Roig, A. (1984) The
influence of heavy metals upon the growth of
sitka-spruce in South Wales forests II.
Greenhouse experiments. Plant and Soil, 78(3),
271-282.
Received: 01.04.2016
Accepted: 11.07.2016
CORRESPONDING AUTHOR
Nevena Cule
Institute of Forestry, Kneza Viseslava 3
11030 Belgrade, Serbia
e-mail: nevena.cule@yahoo.com