Previous PageTable Of Contents

Allelopathy for bio-control of water hyacinth

R.M. Kathiresan

Department of Agronomy, Faculty of Agriculture, Annamalai University, Annamalainagar, Tamilnadu, INDIA- 608 002.



Choice of a single control option has been observed to be inefficient for managing the world’s worst aquatic weed water hyacinth. Integrating several tools, especially those which are quick and complete in action are preferred. Use of herbicides and mechanical clearance that could offer such results are constrained with environmental safety and expensive labour and cost involvement. In this situation, allelopathic potential of some native plant species against water hyacinth proves to offer an effective solution to the problem. Such allelopathic plant materials were chosen through whole plant and incised leaf bio-assay and their allelopathy were confirmed through micropond studies, mechanism of action traced through experiments using leaf discs of water hyacinth. The ecological safety of such plant materials was also studied.

Media Summary

Different allelopathic plant materials have been shown to be successful for selective control of water hyacinth and safe for non-target flora. Their dose response pattern, mode of action are discussed.

Key Words

Allelopathy , water hyacinth , microponds, leafdisc- assays, eco-safety


The invasive alien aquatic weed water hyacinth (Eichhornia crassipes Mart. Solms.Laubach) poses a severe threat in India, where nature’s fury in the form of drought and associated problems are a chronic feature. The weed reduces the volume of available fresh water by evapo-transpiration losses that increase up to 9.84 times. It has been observed to impede the flow of water in irrigation systems, impair the quality of water and favour breeding of vectors of human diseases as reported by Terry (1996). Lake Veerannum and it’s distributaries with a command area of 18,000 ha, forms the major irrigation source of Tamilnadu state in South India. This lake and its distributaries have been severely infested with water hyacinth, in recent years (Kathiresan, 2000). The different management options available have their own merits and constraints. Exclusive dependence on classical bio-control using insects is constrained with the length of time (Harley et al. 1996) and remnants of the weed below threshold level even after completion of the project. Further, in many watershed environs of India, classical bio-control with insects is limited by interrupted host range (weed vegetation) due to drying of water bodies in the summer. As regards herbicide use, toxicity on non-target flora and fauna and residue hazards militate against their adoption. Further, no single herbicide is registered for use on water, in India. Similarly, mechanical weeding is labour and cost expensive. Experience gained in control of water hyacinth in different regions of the world showed that single stroke approach was inefficient in tackling water hyacinth and several tools working together has been observed to be the best alternative (Cordo, 1998; Labrada et al. 1996). Among the short-term control measures, using allelopathic plant materials has been shown to be a feasible option (Kathiresan, 2000). Screening of different alleolopathic plant products for the management of water hyacinth has been taken up at Annamalai University, India. The allelopathic potential of the native plants have been investigated, through a series of Laboratory, Green house and field experiments, for exploitation as components of a water hyacinth management programme.


The study comprised two green house, four laboratory and one field experiment comprising seven experiments in total. Screening of different allelopathic plant products was done initially with whole plants of water hyacinth grown in polystyrene cups of 200ml capacity in the green house. There after a dose response pattern for highly allelopathic plant materials was evolved using incised leaf bioassay in the laboratory. Response of differing growth stages of water hyacinth to these plant materials were later studied in another green house experiment. Final and confirmatory field testing of these plant products were done in micro ponds. The mode of action of the plant materials on water hyacinth were traced through laboratory experiments involving leaf disc. The impact of the allelopathic plant materials on microbes and paddy and cucumber were studied through another set of laboratory experiments. Dried leaf and stem powder of different plants numbering 55 were used in the initial screening. These plant products were used for treating the water hyacinth plant at 30gl-1 and compared with an untreated control. For the sake of space constraint, precision and convenience, only selected plant products and observations are presented in this paper. The water hyacinth weed was observed for biomass reduction, reduction in chlorophyll content and reduction in number of healthy leaves plant-1. The plant products were graded using the results of the initial screening into four categories viz., highly allelopathic, moderately allelopathic, less allelopathic and non allelopathic. Those plant products that were graded as highly allelopathic were subjected to further testing under varying doses with whole plants and cut leaves of water hyacinth for eliciting a dose response pattern. For this cut leaf bio assay, the leaves of the weed (with healthy leaves submerged in water) were incised on the petiole with a razor blade with care to retain the incision point below water level in order to avoid trapping air bubbles in the petiole. Afterwards, the pattern of allelopathic inhibition of three different growth stages of water hyacinth as identified by Kannan and Kathiresan (1998) were also studied in the green house, by subjecting individual plants of these weed growth stages in plastic trays to these plant products. Testing the performance of these plant products under natural aquatic environments was taken up with simulated field conditions in micro ponds. Pits of dimension 0.75 X 0.75 X 0.75 m were dug in the fields, to constitute micro ponds. Channels with running water were provided on either side of these pits to avoid seepage loss of water from the pits. Initially these micro ponds were flooded with water and receding water level was topped up until 20 days. Afterwards, healthy water hyacinth plants were released in these micro ponds and allowed to establish for 2-3 days. Thereafter the plant products were let in @ 30 g l-1. Laboratory experiments were taken up with the leaf discs of water hyacinth to study the mode of action of these plant products. One cm leaf discs of the weed were prepared from whole plants using cork borer and twenty such discs were placed in 50 ml of aqueous extracts of the highly allelopathic plant products. The plant products at doses of 30, 25, 15 and 10 g l-1 were soaked in 200 ml of water for 24 hr and filtered through Whatman No.1 filter paper and made up to 200 ml using distilled water for the study. Conductivity readings using a Radio Copenhagen conductivity meter, were taken immediately (0 hr) and again after 4 and 8 hr and recorded in m.mhos cm-1.These extracts were also incubated for 5, 15 and 30 days to study their impact on micro organisms. They were inoculated at a concentration of 10-2 in Rose Bengal Agar medium and counted for fungal population on 4th day. Inoculating the extracts at a concentration of 10-3 was done in Nutrient Glucose Agar medium and observed for bacterial counts on 3rd day. Viable seeds of paddy and cucumber were soaked overnight and surface sterilized. These seeds were later spread on Petriplates @ 10 per plate over a moist filter paper maintained with a slow and continuous exposure to 10 ml of aqueous extracts of these two allelopathic plant products at respective doses, and 10 ml of distilled water (for comparison as control), observed for germination on 3rd day and radicle and plumule length on 7th day.

Results and Discussion

Screening of plant products for allelopathic control of water hyacinth using the whole plants showed a clear difference in the potential among them in eliciting allelopathic inhibition. Four different ranges in the magnitude of inhibition were observed. As regards the biomass reduction of whole plants, the ranges of inhibitory response were 30 per cent and above, 29 per cent to 15 per cent, less than 15 per cent and no inhibition after 48 hours of exposure. Accordingly, these ranges were used to categorize highly allelopathic, moderately allelopathic, less allelopathic and non-allelopathic groups of plant products as tabulated in another paper entitled “Evaluation of allelopathic plant materials for aquatic weed control”, presented in this congress. Ten out of 55 different plant products including Coleus amboinicus Benth and Parthenium hysterophorus L. were highly allelopathic. Another 12 including Acalypha indica L, Trianthema portulacastrum L and Sesbania grandiflora (L) Pers were moderately allelopathic. Twelve others with Croton sparsiflorus Morong, Cleome viscosa L and Eclipta alba were less allelopathic. The remaining 21 including Leucas aspera spreng, Curcuma longa L and Euphorbia hirta L were non-allelopathic, without showing any allelopathic inhibition over water hyacinth. The differential response of whole plants of water hyacinth to the different plant products varying from highly allelopathic to non-allelopathic could be due to any one or combination of the following factors i) difference in the biochemical nature of the allelocompounds present; ii) number and quantity of allelocompounds involved, iii) persistence of the allelocompounds upon liberation in the system, iv) uptake and translocation of the allelocompounds by the target weed. For example, the highly allelopathic plant product that showed highest activity in the present study was C.amboinicus and the same was shown to be highly allelopathic on E.crassipes and other aquatic weeds including algae (Kathiresan, 2000). P.hysterophorus was also reported to be highly allelopathic on E.crassipes (Pandey et al., 1993). The mechanism of action of parthenin, one of the allelocompound present in P.hysterophorous was reported to be associated with decline in water use, root dysfunction, excessive leakage of solutes from roots, massive damage to cell membranes and loss of chlorophyll (Pandey, 1996). This literature is in conformity with the present investigation. This study also traced higher magnitude of electrolyte leakage from leaf discs of water hyacinth on exposure to these plant products. This shows that the principle mode of action of allelopathic plant products on water hyacinth was membrane disruption and solute leakage. The magnitude of inhibitory response elicited was comparatively higher with cut leaves than with whole plants. This could be attributed to the leaf physiology or system directly getting exposed to the allelocompounds without any scope for the allelocompounds to get metabolized during the process of translocation from the roots (as in case of absorption of the plant products by the whole plants of E.crassipes).

Regarding the dose response studies, bio-mass reduction in incised leaves of water hyacinth after 24 hrs of exposure to the plant products of the three allelopathic categories are furnished in Table 1.

Table 1. Percentage reduction in fresh weight E. crassipes cut leaves after 24 hr of treatment with graded dosages of various allelopathic plant products. Figures in parentheses are original values before arc-sine transformation






30g l-1
(30000 ppm)





25 g l-1
(25000 ppm)





20g l-1
(20000 ppm)





15 g l-1
(15000 ppm)





10 g l-1
(10000 ppm)





5 g l-1
(5000 ppm)





2.5 g l-1
(2500 ppm)





1.0 g l-1
(1000 ppm)





0.5 g l-1
(500 ppm)




















C. amboinicus recorded the highest magnitude of allelopathic inhibition of water hyacinth at all the doses tried ranging from 30,000 ppm to 500 ppm. Though the inhibitory response in all the plant products were observed to be dose dependent, the response of allelopathic inhibition to reducing levels of dose were not proportional, especially with C.amboinicus and P.hysterophorus. When the doses of these plant products were reduced from 30000 ppm by three times to 10,000 ppm, the inhibitory response came down only by around 7 and 10 per cent with C.amboinicus and P.hysterophorus, respectively. Even at the least dose of 500 ppm, these plant products exhibited more than half of the allelopathic inhibition obtained at the highest dose that is 60 times higher. These observations indicate that the semiochemicals present in the plant products either independently or in combined manner injure the weed and their allelopathy is significant even at considerably lesser doses. For practical field uses, the optimum dose need to be selected based on the potential and time taken to inflict lethality, ability of the weed to recover from the damage and their impact on non-target flora and fauna. The differential composition of allelochemicals in some of these plant products as observed earlier and furnished also lend support for these observations. C.amboinicus : α - humulene, carvacrol, thymol α - pinene, α - terpine (Vasque et al., 1999 and Kathiresan, 2000). P.hysterophorus : p-hydroxybenzoic acid, Parthenin, Caffeic acid and p-coumaric acid (Das and Das 1990). L. leucocephala : mimosine, quercetin, gallic, protocatechuic, p-hydroxyphenylacetic, vannilic, caffeic and Coumaric acid (Chou, 1995). A. indica : Tirucallol and azadiractin ( Isman, 1990). It was also observed that immediate response of water hyacinth to allelopathic inhibition varied with differing stature or growth stage. Inhibition was higher with smaller plants and lesser with larger plants (Table 2.)

Table 2. Percentage reduction in fresh weight of different growth stages of E.crassipes due to highly allelopathic plant products. Figures in parentheses are original values before arc-sine transformation






Stage I

30 g l-1 (30000 ppm)

56.65 (69.78)

90.00 (100.00)

90.00 (100.00)

20 g l-1 (20000 ppm)

38.30 (38.41)

90.00 (100.00)

90.00 (100.00)

10 g l-1 (10000 ppm)

13.57 (5.51)

20.41 (12.16)

21.96 (13.99)













Stage II

30 g l-1 (30000 ppm)

40.91 (42.90)

90.00 (100.00)

90.00 (100.00)

20 g l-1 (20000 ppm)

24.57 (17.29)

43.45 (47.30)

90.00 (100.00)

10 g l-1 (10000 ppm

12.45 (4.65)

17.01 (18.56)

18.43 (10.00)













Stage III

30 g l-1 (30000 ppm)

35.05 (32.98)

90.00 (100.00)

90.00 (100.00)

20 g l-1 (20000 ppm)

20.39 (12.13)

36.20 (34.89)

90.00 (100.00)

10 g l-1 (10000 ppm)

12.87 (4.96)

20.25 (11.98)

21.11 (12.98)













However, with prolonged exposure, this variation in inhibition disappeared and both the highly allelopathic plant products imparted complete biomass reduction at higher doses in all the three growth stages of the weed viz. small, medium and large. This might be due to the pace of interruption of physiological functions varying with the stature of the plant, with large sized plants requiring higher quantity of allelochemicals for stalling the higher rate of metabolic functions involved and smaller sized plants requiring lesser quantities of the allelochemicals for the same. With the lapse of time, more of allelochemicals are absorbed by E.crassipes at all growth stages that might have been adequate to knock down infrastructure for any amount of metabolic functions. This study brought out the fact that highly allelopathic plant products at higher concentration viz., 30000 ppm would be sufficient to tackle the weed regardless of its stature. Higher magnitude of inhibition or activity of herbicides on younger plants compared to that of mature plants of E.crassipes reported by Gurwre et al. (1999) lends support for these results of the present study. Results of the micropond studies indicate that among the selected highly allelopathic plant products compared for their activity on E.crassipes under field conditions, C.amboinicus was exhibiting a higher magnitude of inhibition at higher concentrations of 30000 ppm and 20000 ppm. P.hysterophorous was only next in order. C.amboinicus was able to impart a near cent per cent fresh weight reduction of the weed by 4 DAT under both 30,000 ppm and 20,000 ppm concentration (Fig1.), Whereas P.hysterophorous at 20,000 ppm took more than six days to achieve such a perfect weed suppression (only on 8 DAT). This indicates that the combined effect of different allelocompounds liberated from C.amboinicus was able to act better at higher concentrations than that from P.hysterophorus at the same concentration.

Fig 1. Percentage reduction in fresh weight of E.crassipes due to selected highly allelopathic plant products @ 20 g l -1 (20000 ppm) in micro ponds

Fig 2. Bacterial population ml-1 of aqueous extracts of various highly allelopathic plant products (x 103 ) on 5 , 15 and 30 DAI

Leakage of electrolytes form leaf discs of test species is considered to be and indication of disrupted membrane integrity and consequent interruption of related metabolic functions like photosynthesis and respiration (Harper and Balke, 1981). Highly allelopathic plant products were shown to induce leakage of ions or electrolytes from the leaf discs of E.crassipes. Increased electrical conductivity measured in aqueous extracts of these plant products wherein the leaf discs were floated, in comparison with that of distilled water shows that the principle mechanism of action of all these plant products on E.crassipes is through disruption of cellular membrane permeability. Higher magnitude of electrical conductivity is recorded with C.amboinicus at all the doses compared (Table 3.).

Table 3. Effect of allelopathic plant product on electrolyte leakage from leaf discs of E. crassipes




0 hr

4 hr

8 hr

0 hr

4 hr

8 hr

T1- 30g l-1 (30000 ppm)







T2- 25g l-1 (25000 ppm)







T3- 20g l-1 (20000 ppm)







T4- 15g l-1 (15000 ppm)







T5- 10g l-1 (10000 ppm)







T6- control







Similarly, gradation in the electrolyte leakage as measured in terms of electrical conductivity is observed with varying doses of plant products. Higher doses recorded higher electrical conductivity readings than the lower doses (C.amboinicus at 30000 ppm recorded 8.47 m.mhos cm-1 at 4 hr while the same at 10000 ppm recorded 1.94 m.mhos cm-1 only). This adds to the understanding that the quantum of damage interms of membrane disruption and electrolyte leakage is proportional to the quantity of allelocompounds to which the leaf discs are exposed. Further, the increasing magnitude of leakage of electrolyte with increasing time (4 hr and 8 hr) indicates greater damage with longer periods of exposure. These findings corroborate with the observations in the present study regarding initial screening, dose response studies and micropond studies. Earlier workers also reported similar findings (Macdonald et al. 1993). Highly allelopathic plant products were tested for their activity on non-target flora comprising bacteria and fungi. Results showed that both these group of organisms suffered an inhibitory response during the initial stages after treatment viz, 5 DAI and 15 DAI. However, the inhibition on bacterial and fungal counts disappeared at later stages viz., 30 DAI and resilience was observed in respect of highly allelopathic plant products (Fig 2). This could be explained by the fact that allelochemicals also suppress microbes like bacteria and fungi, immediately after release in the ecosystem. However, these microbes after passing through a lag phase develop adaptive strains and build up or rebound in adequate population contributing for the resilience pattern of colonizing. Accordingly, it could be understood that the plant products though of higher inhibitory potential on E.crassipes, might well be degraded easily by bacteria and fungi over time without suffering any sustained or permanent injury. This observation regarding environmental safety with natural products is supported by the earlier reports of Lodhi and Rice (1971). C. amboinicus that recorded enormous and rapid activity on E.crassipes was found to be safe for use in aquatic systems without affecting the germination, plumule and radicle length of paddy as well as cucumber (Fig. 3). All other plant products except P.hysterophorous, D.metel and L.leucocephala were found to be safe for paddy and cucumber as no inhibitory response was shown by the seeds after treatment in terms of germination percentage, plumule and radicle length. P.hysterophorous was found to be allelopathic on paddy as well as cucumber. Their allelopathic inhibition varied with different concentrations, with higher magnitude of inhibition at higher concentrations. This could be due to varying nature of allelocompounds present in the different plant products. One or more of those of the allelochemicals involved in P.hysterophorous might have been allelopathic over paddy and cucumber, whereas such of those allelocompounds present in C.amboinicus might not have been allelopathic over paddy and cucumber. The varying pattern of susceptibility of paddy and cucumber with different allelocompounds could be attributed to differential uptake and metabolism of the compound by the target plant.

Fig 3. Effect of extracts of various highly allelopathic plant products @ 30 g l-1 on Paddy and Cucumber.


Allelopathic potential of some plants offer scope for integrating them as an effective short term bio-control tool for managing water hyacinth, while ensuring safety on non-target flora and fauna.


Author gratefully acknowledges the encouragement and facilities extended by authorities of Annamalai University and the financial support offered by organizers towards his participation in the congress.


Chou H (1995) Allelopathy and sustainable agriculture. Allelopathy organisms, process and applications, (Eds M Chou and P Chen ) pp. 211-223 Intercept Ltd., UK

Cordo HA (1998) New agents for biological control of water hyacinth. In “Proceedings of the I IOBC Working Group Meeting for the biological and Integrated control of water hyacinth, PPRI, Pretoria, South Africa”. (Eds M P Hill Mic Julien and D Center ) pp. 68-74.

Das B, Das R (1990) Chemical investigation in Parthenium hysterophorus an allelopathic compound. Allelopathy Journal 2(1), 99-104

Gurwre RM, Mabasa S, Mazarura U, Mangosho E, Mtsekwa M (1999) Screening of herbicides for their potential to control water hyacinth in Zimbabwe. In “Proceedings of the IOBC Working Group Meeting for the biological and Integrated control of water hyacinth, PPRI, Pretoria, South Africa”. (Eds M P Hill Mic H Jlien and Ted D Center) pp.148-152.

Harley KLS, Julien MH, Wright AD (1996) Water hyacinth : a tropical world wide problem and methods for its control. In “Proceedings of the Second International Weed Control Congress, Copenhagen, Denmark”. pp 639-644.

Harper JR, Balke NE (1981) Characterization of the inhibition of K+ absorption in oat roots by salicylic acid. Plant Physiology, 68, 1349-1357

Isman MB, Koul O, Lucynski A, Kaminski J (1990) Insecticidal and antifeedant bioactivities of neem oils and their relationship of azadiractin content. Journal of Agricultural Food Chemistry, 38,1406-1411.

Kannan, C, Kathiresan RM (1998) Biological Control of Water hyacinth at different growth stages. In “Proceedings of the first meeting of Global Working Group on Integrated and Biological Control of Water hyacinth, IOBC, Harare, Zimbabwe”. pp.1-9.

Kathiresan RM (2000) Allelopathic Potential of Native plants against water hyacinth. Crop Protection. 19(8-10), 705-708. (Elsevier Science, UK)

Labrada R, Charudattan R, Center TD (1996) International expert consultation on strategies for water hyacinth control: Background and Justification. In “Strategies for water hyacinth. FAO report of a panel of experts Meeting, Ft. Lauderdale, Florida, (Eds. Charudattan et al) pp. 1-2.

Lodhi MAK, Rice EL (1971) Allelopathic effects of Celtis laevigata. Bulletin of the Torrey Botanical Club 98, 83-89.

Macdonald G E, Shilling D G, Bowick T A (1993) Effects of endothall and other aquatic herbicides on chlorophyll, fluorescence, respiration and cellular integrity. Journal of Aquatic Plant Management, 31, 51-54.

Pandey D K (1996) Phytotoxicity of sesquiterpene lactone parthenin on aquatic weeds. Journal of Chemical Ecologly, 2(1), 151-160.

Pandey D K Kauraw L P and Bhan V M (1993) Inhibitory effects of parthenium (Parthenium hysterophorus L) residue on growth of water hyacinth (Eichhornia crassipes (Mart.) Solms). Journal of Chemical Ecology, 19(11), 2663-2670.

Terry PJ (1996) The water hyacinth problem in Malavi and foreseen methods of control. In “Strategies for water hyacinth. FAO Report of a Panel of Experts Meeting, Ft. Lauderdale, Florida,USA.” (Eds. Charudattan et al.) pp. 59-81.

Vasquez EA, Morallo Rejesus B, Punzalan E G, Kraus W (1999) Biological studies on the essential oil of Coleus amboinicus. In “Abstracts of the XIV International Plant Protection Congress, Jerusalem, Israel.” p.29.

Previous PageTop Of Page