Intake, nutrient digestibility, rumen fermentation and water kinetics of sheep fed Dichanthium annulatum grass hay-tree leaves diets

Sultan Singh and S S Kundu*

Plant Animal Relationship Division, Indian Grassland and Fodder Research Institute, Jhansi-284003, India
singh.sultan@rediffmail.com
* National Dairy Research Institute, Karnal, India

Abstract

In 4 separate feeding trials, 4 adult male Muzzafarnagari sheep (mean weight of 26.8± 2.0 kg) were fed mixtures of Dichanthium annulatum (DA) grass and leaves of Helictris isora (HI), Securenegia virosa (SV), Leucaena leucocephala (LL) and Hardwickia binnata (HB), in 75:25 ratios (DM basis) for 90 days on each diet. Feed intake, nutrient utilization, rumen metabolites and rumen water kinetics were measured.

LL had highest CP concentration (20.6%) and HB the lowest CP (11.2%) of the tree leaves, while grass ranged from 4.4 to 5.4% across the feeding trials. Dry matter intake was higher (P<0.05) on DA-HI (3.3% bodyweight and 74.0 g/kgW^0.75) than on other diets. DM, CP, NDF, ADF and cellulose digestibility was higher (P<0.05) on DA-HI and DA-SV than DA-HB diet. Faecal N loss was (P<0.05) higher on DA-LL and DA-HB, while urinary N excretion was (P<0.05) more on DA-HI diet. Sheep was in negative N balance on DA-HB diet. Meal size (g/h, % body wt/h and g/kgw^0.75/h) was higher (P<0.05) on DA-HI and DA-SV than DA-HB diet. Eating rate (g/h) was higher on DA-HI diet and lowest on DA-HB for 1^st and 4^th h of eating. In the last 7 and 17 h of feeding sheep eating rate was more on DA-LL and lowest on DA-HB.  TVFA and N metabolites concentration was higher (P<0.05) on DA-LL than other diets. Out flow rate (l/d) was more in sheep on DA-HI than other diets, while rumen volume was comparable (P>0.05) on diets.

Results revealed that sheep had higher intake, more nutrients digestibility and optimum rumen metabolites concentration on DA-HI and DA-SV diets while feeding of DA-HB diet lowered nutrients digestibility and rumen metabolites production in sheep

Key words: eating pattern, nutrients digestibility, tree leaves rumen metabolites, rumen water kinetics, sheep

Introduction

Grasses with leaves of browse plants (shrubs/trees) constitute a major component of small ruminant feeding systems both under grazing and semi-intensive systems in subtropical and tropical regions (Devendra 1990; Topps 1992). Leaves of many trees and shrubs are rich sources of protein, vitamins and minerals (Aganga and Mesho 2008; Mtui et al 2008) for supplementing low quality roughages (grasses, straws and stovers) (Aganga 2003).

Availability, nutritional quality, animal species and concentrations of toxic compounds restrict the level of use of tree leaves in roughage-based diets for small ruminants (Devendra 1990;  Singh 2004), although these authors recommended the feeding of shrub and tree leaves at 30-50% of the diet (DM basis). Presence of toxic and secondary metabolites limits the use of browse species in animal feeding (Silanikove et al 1997; Getachew et al 2002). These compounds exert both beneficial and harmful effects on ruminant digestive physiology (rate of passage, rumen metabolism, and microbial activity) and efficiency of use of nutrients viz. intake and digestibility (Waghorn et al 1994; Mc Sweeny et al 2001). Responses in sheep rumen fermentation and digesta kinetics from supplementing Pennisetum hay with tree leaves have been demonstrated (Navas-Camacho et al 1993).     

Evaluation of locally available feed resources including tree leaves/shrubs to reduce the feed deficit is of paramount importance, as they constitute a sizeable part of livestock feeding, particularly for small ruminants. While the nutritive value of Helictris isora, Securenegia virosa, Leucaena leucocephala and Hardwickia binata tree species available in Bundelkhand region, a part of semi-arid India, has been evaluated (Negi et al 2003; Chaurasia et al 2006), no information exists on the effects of Helictris isora, Securenegia virosa and Hardwickia binata leaves on rumen function.   The present study aimed to assess the effects on intake, nutrient digestibility, rumen fermentation and rumen water kinetics in sheep supplemented with leaves of these trees when fed Dichanthium annulatum grass-based diets.

Materials and methods 

Collection of grasses, tree leaves and shrubs samples

Tree leaves were collected by chopping the small tender branches and twigs from the trees of grazing fields and nursery of Grassland and Silvipastoral Management Division of the Institute. The leaves were dried under shade on polythene sheets for later use in animal feeding. Dichanthium annulatum (DA) grass was harvested at full bloom to mature stage from its pure fields maintained at Grassland and Silvipastoral Management Division and conserved as hay.

Animals and diets

Four adult male Muzzafarnagari sheep selected from the flock maintained at Plant Animal Relationship Division of Institute were offered Dichanthium annulatum-Securengia virosa (DA-SV), Dichanthium annulatum-Helictris isora (DA-HI), Dichanthium annulatum-Leucaena leucocephala (DA-LL) and Dichanthium annulatum-Hardwickia binata (DA-HB) diets in a 4 x 4 switch over experiment. Average weight of sheep was 26.7±2.7, 26.5± 2.6, 25.6±2.5 and 28.2±2.6 kg on DA-SV, DA-HI, DA-LL and DA-HB diets, respectively.  Sheep were offered grass and tree leaves in 75:25 ratios (DM basis) in each feeding trial for more than 90 days.  Animals were tied with nylon rope/iron chain individually and offered half each of the grass and tree leaves between 09.00 and 09.30 h and the remaining half in the evening between 15.00 and 15.30 h. The animals were watered twice at 11.00 h and 15.00 h and were maintained under hygienic conditions throughout the experimental period.

Eating pattern

Prior to digestibility trial (after 50-55 days feeding) eating pattern of sheep on DA-HI, DA-SV, DA-LL and DA-HB diets was recorded for 3 consecutive days by offering half quantity each of grass and tree leaves at 09.00 h and again at 13.00 h. The intake of feed by individual animals was recorded at 1 h, 2 h, 3 h, 4 h, 5 h, 7 h and 17 h after the initial feeding by weighing the remaining feeds (refusal/leftover).

Digestion/metabolism trial

A digestion/metabolism trial of 7 days duration was conducted on each diet after about 60 days of feeding. Animals were kept in metabolism cages and allowed to adjust to the cages for 2-3 days prior to actual sampling of faeces and urine.  Faeces and urine of individual animal were collected for 24 hours and pooled in polythene bags and cans, respectively. Representative samples of faeces for DM (1/50 aliquot) and N (1/100 aliquot preserved in 20% H₂SO₄) estimation were collected for individual animals during the trial. A fixed volume (5 ml) of urine was pooled in digestion flasks (containing concentrated sulphuric acid) for individual animals during the collection period. Samples of grass, tree leaves and residue were also collected daily and representative samples were kept for DM estimation. Dried samples of faeces, feeds offered and residues were ground through a 1-mm sieve. Ground samples were stored in plastic containers and used for further chemical and biochemical estimations.

Rumen liquor collection

Rumen liquor samples were drawn from individual animals after the digestibility trial (60 days of feeding) and after 90 days of feeding. On each occasion, about 50-60 ml of rumen liquor was collected before feeding and 4 h post feeding into a 0.5 L pre-warmed thermos using a perforated plastic stomach tube using light suction. Rumen liquor pH was recorded immediately using the digital pH meter (Systronic-310). Rumen liquor was strained through double layer of muslin cloth and stored in 50 ml plastic bottles after adding a few drops of saturated mercuric chloride and then preserved in freezer until rumen metabolite analyses.

Water kinetics

Rumen volume, outflow rate and dilution rate of sheep on different DA-tree leaf diets were determined by the method of Smith (1959) and Hyden (1961). PEG-6000 dissolved in luke-warm water was infused into the rumen with a stomach tube and rumen liquor samples were collected subsequently.

Analytical techniques

Samples of grass (DA) and tree leaves (LL, HI, SV and HB), residues and faeces were analyzed for CP, DM and ash (AOAC 1990). The neutral detergent fiber (NDF- without amylase addition), acid detergent fiber (ADF), cellulose and lignin were analyzed as per the procedure of Van Soest et al (1991). Strained rumen liquor samples collected at 0 and 4 h post feeding were analyzed for total nitrogen and ammonia-nitrogen by the methods of McKenzie and Wallace (1954) and Conway (1962), respectively. The protein nitrogen was estimated by the method of McKenzie and Wallace (1954) as TCA-precipitable nitrogen.  Total volatile fatty acids (TVFA) were assayed by the method of Briggs et al (1957).

Statistical analysis

For statistical analysis of data on intake, nutrient digestibility, N balance, eating pattern and rumen metabolites, the General Linear Model (GLM) of SPSS version 12.0 was used with multivariate analysis. The difference in means was tested using Duncan’s Multiple Range Test as per Snedecor and Cochran (1968).

Results  

Chemical composition

The CP content of the DA grass ranged from 4.4-5.4 % during the feeding trials, while that of HI, SV, LL and HB were 12.6, 12.6, 20.6 and 11.2 %, respectively (Table 1). Cell wall polysaccharides (NDF, ADF and cellulose) were higher in HB than in the other tree leaves.

Feed intake, nutrient digestibility and eating patterns

Dry matter intake was higher (P<0.05) in sheep fed on DA-HI (3.28% and 74.0 g/kgW^0.75) than on other diets (Table 2).

However, the intake of sheep on DA-LL tended to be higher (P>0.05) than on DA-SV and DA-HB.  Digestibility of DM, OM, NDF, ADF and cellulose on DA-LL and DA-HB were lower (P<0.05) than on DA-HI and DA-SV, and CP digestibility was lower (P<0.05) on a DA-HB diet than on DA-HI and DA-LL.

Consumption of feed on fresh and DM basis (g/h, % body wt g/h and g/Kg W^0.75/h) during 24 of eating was lower (P<0.05) on DA-HB than other diets (Table 3).  

Sheep exhibited higher (P<0.05) eating rate during 1^st and 4^th h of feeding on DA-HI and DA-SV than DA-HB while during 7^th and 17^th h eating rate was not affected by the diets but tended to be higher with DA-HB (Figure 1).

Nitrogen balance

Nitrogen intake (g/d) was higher (P<0.05) in sheep on DA-LL than on other diets (Table 4) and faecal nitrogen excretion was higher (P<0.05) on DA-LL and DA-HB than on the other diets.

Urinary nitrogen loss on DA-HI was higher (P<0.05) than on the other diets. Animals were on negative nitrogen balance on DA-HB diet (-0.2 g/d), while N balance was highest on DA-LL diet (2.7 g/d).

Rumen metabolites and water kinetics

The mean pH of rumen liquor was similar on all diets (Table 5), though it was lower at 4 h after feeding than before feeding on all the diets.  

Concentrations of total volatile fatty acids, total N, ammonia nitrogen and TCA soluble N in sheep rumen liquor were significantly (P<0.05) higher on DA-LL than on other diets.

Sheep rumen volume was identical across DA-tree leaves diets (5.6-6.1 Table 5), however dilution rate (%) and outflow rate (l/d) tended to be higher on DA-HI (7.54 and 9.7) than other diets.

Discussion       

Intake, nutrients digestibility and eating pattern

The low CP and high fiber concentrations in the grass reflect the maturity of the grass at harvesting. DMI of 62.4 and 58.4 g/kg w^0.75 in goat fed timothy hay-mulberry leaves in 75:25 and 50:50 ratios, respectively observed by Enzewa et al (2000) is comparable with our observations. Feed intake recorded by Ebong (1995) in range of 69.0 –70.2 g/kg w^0.75 in sheep fed tef straw (Eragrostis tef ) supplemented with A nilotica, A seyal and S sesban is in conformity with our results. Lower feed intake of sheep on DA-HB diet may be partly explained by the higher cell wall contents of HB as proportion of cell wall in a feed/diet accounts for a large amounts of the variation in its intake (Waldo 1986). Meissner and Esterhuyse (1993) showed that DM, N, ash and cell wall constituents explain 70 % of the variation in forage intake.

Lower nutrients digestibility in sheep on DA-HB diet may be attributed to higher fiber and lignin contents of HB, while for DA-LL diet more phenolics and condensed tannin in LL than HI and SV (Negi et al 2003) might have reduced the digestibility by binding with protein and carbohydrates (Leinmuler et al 1991; Hove et al 2001). Chemical entities of a feed and fodder influence their extent of digestion (Van Soest 1994). Alike our findings Hove et al (2001) observed variability in nutrients digestibility in goat on supplementing native pasture hay with dried leaves of legume shrubs.

More fibrous leaves of HB might have resulted in rumen fill and more time in grinding and chewing that might have caused low intake in early hours of feeding. Extent of eating pattern and feed consumption are a function of the type of feed/fodder, physical and chemical properties. Quality of feed has a greater proportional effect on eating rate than on the total amount of feed eaten.   Frisch and Vercoe (1969) reported that eating pattern of animals is highly correlated with fasting metabolism and intake both within and between diets. Without choice of food, taste and sense play a major role in initiation of feeding than amount of feed eaten (Balch and Campling 1962).

Nitrogen balance

Feacal and urinary N excretion in sheep on different DA grass tree leaves diets was lower than the observations of Navas-Camacho et al (1993) and this may be due to low nitrogen intake in present study. Alike the pattern of feacal and urinary N excretion of our study Hindrichsen et al (2004) also recorded varying level of feacal and urinary N excretion in sheep fed maize stover supplemented with different tree foliage.  There is direct relation between the dietary protein level and amount of N excreted in faeces. Nitrogen retention in the animal depends on its excretion through faeces and urine. Higher urinary N excretion in HB-HI diet is not known but it may be due to lack of adequate soluble carbohydrates for microbial protein synthesis as the fiber digestibility was low in sheep fed DA-HB diet. Further the variability in N excretion in faeces of sheep fed DA grass-tree leaves based diets may be associated with the level of tannins in tree leaves (Mc Brayer et al 1983 and Negi et al 2003). Nitrogen retention (g/d) in goats fed pasture hay with L leucocephala and A angustissima and Calliandra calorythus at 80, 160 and 320 g/d level supplementation (Hove et al 2001) follows the similar pattern of present study. Alike the present study Ebong (1995) found (P<0.05) variability in the faeces and urinary N excretion in sheep fed tef straw with different tree leaves.

Rumen metabolites and water kinetics

Ebong (1995) also observed significant effect (P<0.01) of tree leaves (A seyal, A nilotica and S sesban)  supplementation to tef straw on rumen liquor TVFA and NH₃-N of sheep like present but the NH3-N level was low while TVFA at par with our results.. Variation in chemical composition of tree leaves might be responsible for differences in N degradability (Fall Toure and Michalet-Doreau 1995) resulting in variable NH3-N concentration in sheep rumen liquor on evaluated diets. Level of N metabolites in rumen liquor depends on the protein content and its soluble fractions. High contents of soluble protein (20-60%) may be found in high protein feeds such as Leucaena, Gliricidia and Calliandra (Preston and Leng 1987). Higher protein content and more solubility of LL protein may be responsible for higher contents of N metabolites in sheep on DA-LL diet. The NH₃-N concentration for all the diets was above the minimum concentration of 50mg/l for microbial growth (Satter and Slyter 1974) and 23.8 mg/100 ml for optimum microbial fiber digestibility (Mehrez et al 1977). The concentration and composition of tannins and phenols present in shrubs and tree leaves inhibit the fermentation of protein and carbohydrates (Woodward and Reed 1989; Kibon and Orskov 1993) resulting in variation in rumen metabolite production on different diets. Rumen volume (L) of sheep fed DA-tree leaves diets was relatively less than reported by Navas-Camacho et al (1993) on Pennisetum clandestinum grass and Enterolobium ciclocarpum at 100 and 300 g of supplementation, however the dilution rate (%/h) was comparable to present observations. The decreased dilution rate (4.2 %/h) on DA-HB diet should have increased the retention time of solids resulting in more DM degradation. But this did not happen in present study like Navas-Camacho et al (1993). Manyuchi et al (1997) observed no effect of napier or groundnut hay supplementation to pasture hay fed to sheep on rumen volume but observed an increase of the out flow rate.

Conclusions 

• Sheep had higher feed intake and nutrients digestibility on Dichanthium annulatum-Helictris isora (DA-HI)  and Dichanthium annulatum-Securengia virosa (DA-SV) diets, while rumen metabolites and N retention wee higher with  Dichanthium annulatum-Leucaena leucocephala (DA-LL). Hardwickia binata is not a good supplement to DA grass, as on DA-HB diet there was low feed intake, nutrients digestibility and rumen metabolites production.

Acknowledgements 

Financial support under Young scientist Scheme by Indian Council of Agricultural Research, New Delhi is highly acknowledged. Authors are also thankful to Director IGFRI, Jhansi for providing facilities to carry out the work. 

References 

Aganga A A 2003 Indigenous browses as feed resources for grazing herbivores in Botswana. African Journal of Science Technology Science Engineering Series 3: 14-19

Aganga A A and Mesho E O 2008 Mineral contents of Browse plants in Kweneng district of Botswana. Agricultural Journal 3 (2): 93-98 http://docsdrive.com/pdfs/medwelljournals/aj/2008/93-98.pdf

AOAC 1990  Official Methods of Analysis. 13th Edition. pp. 69-70. (Association of Official Analytical Chemists, Washington (DC)

Balch C C and Campling R C 1962 Regulation of voluntary feed intake in ruminants. Nutritional Abstracts and Reviews 32: 669-686

Briggs P K, Hogen J P and Reid R I 1957 Effect of volatile fatty acids, lactic acid and ammonia on rumen pH in sheep. Australian Journal of Agricultural Research 8: 674

Chaurasia M, Kundu S S, Singh Sultan and Mishra A K 2006 Cornell net carbohydrate and protein system for nutritional evaluation of tree leaves, shrubs and grasses. Indian Journal of Animal Sciences 7: 81-87     

Conway E J 1962 Micro-diffusion analysis and volumetric error, fifth edition. Lockwood and Sons Ltd, London

Devendra C 1990 The use of shrubs and tree fodder by ruminants. In shrubs and tree fodders for farm animals (editor C Devendra),pp. 42-60.  Denpasar, Indonesia, 24-29 July 1989, IDRC Ottawa Canada. http://idl-bnc.idrc.ca/dspace/bitstream/10625/19718/1/49268_p221-236.pdf

Ebong C 1995 Acacia nilotica, Acacia seyal and Sesbania sesban as supplements to tef (Eragrostis tef) straw fed to sheep and goats. Small Ruminant Research 18: 233-238

Enzewa I, Kitachar N, Nishida T and Shibata S 2000 Intake and digestibility in Sannen goats fed timothy hay supplemented with mulberry leaves. Proceedings 7^th International Congress on Goats, pp. 127. Tours, France.

Fall-Toure S and Michalet-Doreau B 1995 Nitrogen partition in cell structures of tropical browse plants compared with forages: influence on their in situ degradation. Animal Feed Science and Technology 51: 65-72

Frisch J F and Vercoe J E 1969 Live weight gain, food intake and eating arte in Brahman, Africandor and Shorthorn and Hereford cattle. Australian Journal of Agriculture Science 20: 1189-1195

Getachew G, Makkar H P S and Becker K 2002 Tropical browse: contents of phenolic compounds, in vitro gas production and stoichiometric relationship between short chain fatty acid and in vitro gas production. Journal of Agricultural Science 139: 341-352

Hindrichsen I K, Osuji P O, Odenyo A A, Madsen J and Hvelpluned T 2004 Effect of supplementation of maize stover with foliage of various multipurpose trees and Labalab purpureus on intake, rumen fermentation, digesta kinetics and microbial protein supply of sheep. Animal Feed Science and Technology 113: 83-96

Hove  L, Topps J H, Sibanda S and Ndlovu  L R 2001 Nutritional intake and utilization by goats fed dried leaves of shrub legumes Acacia angustissima, Calliandra calothyrus and Leucaena leucocephala as supplement to pasture hay. Animal Feed Science and Technology 91: 95-106

Hyden S 1961 The use of reference substances and measurement of out flow rate in alimentary tract. In Digestive Physiology and Nutrition of Ruminants, pp. 37.  Butterworth, London.

Kibon A and Orskov E 1993 The use of degradation characteristics of browse plants to predict intake and digestibility by goats. Animal Production 57: 247-251

Leinmuler E, Steingass H and Menke H 1991 Tannins in ruminant feedstuffs. Animal Research and Development 33: 9

Mac Brayer A C, Utley P R, Lowrey R S and Mc Cormick 1983 Evaluation of peanut skins (Testa) as feed ingredient for growing-finishing cattle. Journal of Animal Science 56: 173-182 http://jas.fass.org/cgi/reprint/56/1/173.pdf

Manyuchi B, debHavell F D, Ndlovu L R, Topps J H and Tigere A 1997 Napier of groundnut hay as supplement in diets of sheep consuming poor quality natural pasture hay.2. Effect on intake and rumen digesta kinetics. Livestock Production Science 49: 43-52

McKenzie H A and Wallace H S 1954 The kjeldahl determination of nitrogen. Australian Journal of Chemistry 7: 55-61

Mc Sweeny C S, Palmer B, Mc Niell D M and Krause, D O 2001 Microbial interactions with tannins: nutritional consequences for ruminants. Animal Feed Science and Technology 91: 83-93

Mehrez A Z, Orskov E R and Mc Donald I 1977 Rates of rumen fermentation in relation to ammonia concentration. British Journal of Nutrition 38:437 http://journals.cambridge.org/action/displayFulltext?type=1&fid=840656&jid=BJN&volumeId=38&issueId=03&aid=840648

Meissner H H and Esterhuyse F E 1993 Index of forage intake by cattle and compositional factors limiting intake. In VII World Conference on Animal Production, pp.108, Edmonton.

Mtui D J, Lekule F P, Shem M N, Hayashida M and Fujihara T 2008 Mineral concentrations in leaves of nine browse species collected from Mvomero, Morogoro, Tanzania. Journal of Food, Agriculture and Environment 6 (3 and 4): 226-230

Navas-Camach A, Laredo M A, Cuesta A, Anzola H and Lean J C 1993 Effect of supplementation with tree legume forage on rumen function. Livestock Research for Rural Development 5 (2)  1-13 http://www.lrrd.org/lrrd5/2/navas.htm

Negi A S, Bhadoria B K, Nag S K and Kundu S S 2003 Chemical composition of some brose species of Bundelkhand. Animal Nutrition and Feed Technology 3: 53-59

Preston T R and Leng R A 1987 Matching Ruminant Production Systems with Available Resources in the Tropics and Subtropics. Penambul Books Ltd, Armidale NSW Australia.

Satter L D and Slyter L L 1974 Effect of ammonia concentration on rumen microbial protein production in vitro. British Journal of Nutrition 32: 199 http://ddr.nal.usda.gov/bitstream/10113/357/1/CAIN749093501.pdf

Silanikove N, Gilboa N and Nitsan Z 1997 Inter-actions among tannins, supplementation and polyethylene glycol in goats given oak leaves: effects on digestion and food intake. Animal Science 64: 479-483

Singh S 2004 Metabolic and Rumen Microbial Activities in Sheep and Goats on grass-tree leaves/shrubs dietary regimens. Technical Bulletin. Indian Grassland and Fodder Research Institute, Jhansi, India.

Smith R H 1959 The development of function of the rumen in milk fed calves. Journal of Apicultural Science 52: 72-78

Snedecor G W and Cochran W G 1968 Statistical Methods, eighth edition. Iowa State University Press, Iowa, USA.

Topps J H 1992 Potential, composition and use of legume shrubs and trees as fodder for livestock in the tropics. Journal of Agriculture Science (Cambridge) 118: 1-8

Van Soest PJ 1994 Nutrition Ecology of Ruminants, second edition. Cornell University Press, Ithaca, NewYork.

Van Soest P J, Robertson J B and Lewis B A 1991 Methods for dietary fibre, neutral detergent fibre and non starch polysaccharides in relation to animal nutrition. Journal of Dairy Science 74: 3588-3597  http://jds.fass.org/cgi/reprint/74/10/3583

Waghorn G C, Shelton I D, Mcnabb W C and Mccutcheon S N 1994 Effect of condensed tannins in Lotus pedunculatus on the nutritive value of pasture for sheep. 2. Nitrogenous aspects. Journal of Agricultural Science Cambridge 123: 109-119

Waldo D R 1986 Effect of forage quality on intake and forage concentrate interactions. Journal of Dairy Science 69: 617-623 http://jds.fass.org/cgi/reprint/69/2/617

Woodward A and Reed JD 1989 The influence of polyphenolics on the nutritive value of browse: a summary of research conducted at ILCA, ILCA Bulletin 35: 2-11  http://www.ilri.org/InfoServ/Webpub/fulldocs/Bulletin35/influence.htm

Received 6 June 2010; Accepted 17 June 2010; Published 1 August 2010

Go to top