A study was undertaken to determine the reason for the differences in the estimation of pasture mass between dairying research centres in New Zealand and to define a common system of estimation that would overcome these differences and be able to describe the pasture situation on commercial dairy farms in different parts of New Zealand. Differences in the visual estimation of pasture between a standard observer and a local observer at the different dairying research centres in the order of 900 kg DM/ha were identified. This was attributed to a lack of regular visual calibration of persons who regularly assessed pasture, as differences between centres in the actual harvested DM yields of the visually assessed plots varied less (200-400 kg DM/ha). To achieve uniformity it is recommended that all centres measure total herbage mass (the amount of herbage above ground level) to regularly calibrate pasture assessment in dairy grazing management studies and on dairy farms. A reasonably consistent estimation of average farm cover, the ranking of paddocks (r2 > 0.9) for the purposes of determining grazing order and the estimation of pre- and post-grazing herbage mass was achieved using the rising plate meter and L'Huillier & Thomson's standard set of calibration equations published in 1988. With these recommendations, estimation of pasture will be more uniform. Standard calibration of the plate meter for the determination of rate of DM disappearance (pre-post grazing) achieved less consistency. Keywords: herbage mass, pasture assessment, pasture height, rising plate meter, visual pasture estimation

Spring grazing managements which allowed some early inflorescence development, followed by hard grazing in order to remove the reproductive stems (designated ‘late control’), were compared with a conventional close grazing strategy (‘early control’) in three paddock-scale complete randomized block studies with dairy cows grazing perennial ryegrass/white clover pastures during consecutive years (1990–1993) in Palmerston North, New Zealand. Experiment 1 was designed to evaluate the timing of late control, Expt 2 to investigate the most suitable intensity of control, and Expt 3 tested different practical approaches to carrying out the late control management.Late control generated an average herbage production increase of 24% (750 kg DM/ha) in October and November (spring) and 22% (950 kg DM/ha) from January to April (summer–autumn) compared with the conventional management. The increase in herbage accumulation during spring was a consequence of greater perennial ryegrass reproductive development. Increased summer–autumn herbage accumulation after late control management was attributed to enhanced tillering activity and accumulation of perennial ryegrass. Effects on white clover production were inconsistent. Experiment 3 compared the effects of two practical late control strategies, involving either a short period of rapid rotational grazing or a simulation of increased stocking rate normally associated with forage conservation, on the milk production and forage intake of groups of cows in self-contained farmlet systems. During the control period, the reproductive state of the sward caused a small decrease in herbage digestibility. However, the extra herbage produced subsequently did not differ in digestibility from that produced on the conventionally managed sward and resulted in an increase of around 11·5% in total milk solids yield (fat, protein and lactose) per cow, particularly towards the end of the season (March and April).The late control spring grazing management strategy is a viable option for dairy farmers. It should coincide with the time of anthesis of ryegrass plants (late November–early December), and is more effectively achieved when forage conservation practices are used to augment grazing pressure in the grazed area of the farm.

This paper reports the results of a dairying systems study at No. 4 Dairy, Massey University, investigating whether the benefits of contrasting spring grazing managements previously measured in small-scale experiments could be measured within the management constraints of a selfcontained farm production system. Management contrasts were early control (EC), in which pastures were closely controlled throughout spring and summer, with average pasture cover kept at approximately 2000 kg DM/ha, and late control (LC), in which average pasture cover was increased to 2700 kg DM/ha, allowing some reproductive growth through October and November before returning average pasture cover to 2000 kg DM/ha in December. Average pasture cover and pre- and post-grazing cover differences between treatments were achieved over late spring in all three years. However, average pasture cover during summer did not differ between treatments. Milksolids (MS) production per cow during the spring phase was higher for the LC treatment than for EC, but the differences disappeared over the December control phase and overall differences were small and inconsistent. The use of mechanical topping in December 1995/96 reduced the penalty to LC MS production experienced in previous years. Late control spring grazing management did not significantly increase pasture or MS production within the confines of a closed production system, contrary to the results of previous small-plot and paddock-scale experiments. Keywords: dairy cow, dairy systems, Lolium perenne, milksolids production, pasture production, spring grazing management, Trifolium repens

Discrepancies in phase two and three studies can result in significant patient and financial burden, as well as the nonapproval of potentially efficacious drugs. We aimed to determine whether this discrepancy exists for clinical trials in inflammatory bowel disease (IBD). Electronic databases (MEDLINE and Embase) and clinical trial repositories were searched from 1 January 1946 to 12 March 2021, for paired phase two and three studies of advanced therapies for Crohn’s disease and ulcerative colitis. The primary outcome was to compare clinical remission rates between paired phase two and three studies for Crohn’s disease and ulcerative colitis. Multivariable mixed-model meta-analysis was performed to calculate odds ratios (OR) with 95% confidence intervals (CI). The Cochrane risk-of-bias tool was used to grade the risk of bias. Of 2642 studies, 29 were included. Fifteen were phase three, 11 were phase two, one was phase one/two, and two were phase two/three. There were no differences in clinical remission rates between phase two and three studies for Crohn’s disease (OR, 1.07; 95% CI, 0.86–1.34; P = 0.54) and ulcerative colitis (OR, 0.81; 95% CI, 0.48–1.36; P = 0.43). Furthermore, there was a lack of any appreciable differences in study characteristics, inclusion criteria and patient demographics among paired phase two and three studies. Most studies were considered low risk of bias. Overall, paired phase two and three studies demonstrate similar clinical remission rates for advanced therapies in IBD. Whether this applies to newer outcomes, such as endoscopic and mucosal healing remains to be determined.