101 Abstract Browse blocks were developed on wet rush-infested areas of Riverside Farm, near Masterton, by planting 6,000 willow stakes/ha over the winter periods of 2000, 2001 and 2002. This was followed by light grazing in the first year after planting and then more intensive grazing in later years. After 18 months the areas dried out, due to evapotranspiration from the trees, and a herbage sward developed from volunteer species. The quality of the herbage progressively improved over a 4 year period and the browse blocks now support year-round grazing by sheep. In Experiment 1, ewes that grazed willow browse blocks for 86 days in 2004 (including mating) had higher reproductive performance than ewes that were mated on short, drought affected pasture. However, reproductive performance was lower than that from ewes mated on a sole diet of long herbage similar to that growing in the browse blocks. It was calculated that 15 and 65% of the metabolisable energy and condensed tannins consumed by ewes grazing browse blocks was provided by willow trees. In Experiment 2, undrenched lambs grazing either control pasture or browse blocks for 98 days in 2005 had lower liveweight and carcass gains than lambs regularly drenched with anthelmintic. Lambs grazing browse blocks consistently had lower dag scores and undrenched lambs grazing on browse blocks had reduced worm burdens of some economically important internal parasites compared with undrenched lambs grazing pasture. It was concluded that the benefits of browse blocks could be due to both direct and indirect effects and that any benefits in internal parasite management may be accompanied by reduced rates of carcass growth. Guidelines are presented for the grazing management of willow browse blocks over a complete 12 month period. Keywords: willow (Salix), fodder tree, drought, internal parasites. Introduction Supplementary willow, comprising leaves and stems <5 mm diameter, contains approx. 10 MJ/kg dry matter (DM) of metabolisable energy (ME), a total N of 25 g/kg DM and 45 g condensed tannin (CT)/kg DM, making it Willow browse blocks for drought protection and for sustainable management of internal parasites T.N. BARRY 1 , E.L. MCWILLIAM 1 , D.W. PITTA 1 , C.M. DIAZ LIRA 1 , P.D. KEMP 1 , W.E. POMROY 1 , N. LOPEZ-VILLALOBOS 1 , P. GAWITH 2 and G.B. DOUGLAS 3 1 Massey University, Palmerston North 2 Waikoukou, RD 4, Masterton 3 AgResearch, Palmerston North T.N.Barry@massey.ac.nz a suitable supplement for grazing livestock (McWilliam et al. 2005a). In a series of three experiments, McWilliam et al. (2005b) showed that supplementation of ewes grazing low quality drought pasture with willow or poplar cuttings during mating increased lambs weaned/100 ewes mated from 102 to 122. However, mechanically pruning willow/poplar fodder incurred a considerable labour cost, so willow browse blocks were developed, with the objective of livestock doing the harvesting by grazing, so eliminating this labour cost. A first objective was to evaluate browse blocks as an alternative to drought pasture for grazing ewes during mating. In a review, Min et al. (2003) showed that forages containing CT could be used in some circumstances to reduce burdens of internal parasites in grazing sheep. A second objective was to investigate if grazing on browse blocks could be used as part of a sustainable strategy to reduce parasite problems in grazing lambs, so reducing the requirement for anthelmintic drenches. Materials and Methods Development and management of browse blocks Five willow browse blocks totalling 6 ha were planted at 6,000 stems/ha on Massey University’s Riverside Farm in the Wairarapa, approx. 10 km north of Masterton. Planting was done over June/July in 2000, 2001 and 2002. Rush-infested wet areas, that produced no edible forage in the undeveloped state, were selected for development as browse blocks. Generally two and sometimes three hillsides drained into these areas. These areas were mechanically mowed in March to remove the rushes and then blanket sprayed with glyphosate 6 weeks later. Following spraying, some of the areas were then ripped to break up the soil prior to planting. Stakes 70 cm long were inserted into the ground up to their mid point, at 1.2 m intervals. The willow clones used were Tangoio and Moutere (one browse block only) (Salix matsudana x alba). Soil pH, Olsen P (μg P/g) and sulphate (μg S/g) were respectively 5.6, 15 and 10 for the willow browse blocks and 5.5, 30 and 12 for the control pasture areas. Browse blocks were rotationally grazed with experimental sheep from 25 February to 14 May 2004 in