Effect of replacing grass silage with maize silage in the diet on bovine milk fatty acid composition

Even though extensive research has examined the role of nutrition on milk fat composition, there is less information on the impact of forages on milk fatty acid (FA) composition. In the current study, the effect of replacing grass silage (GS) with maize silage (MS) as part of a total mixed ration on animal performance and milk FA composition was examined using eight multiparous mid-lactation cows in a replicated 4 × 4 Latin square with 28-day experimental periods. Four treatments comprised the stepwise replacement of GS with MS (0, 160, 334 and 500 g/kg dry matter (DM)) in diets containing a 54 : 46 forage : concentrate ratio on a DM basis. Replacing GS with MS increased (P < 0.001) the DM intake, milk yield and milk protein content. Incremental replacement of GS with MS in the diet enhanced linearly (P < 0.001) the proportions of 6:0-14:0, decreased (P < 0.01) the 16:0 concentrations, but had no effect on the total milk fat saturated fatty acid content. Inclusion of MS altered the distribution of trans-18:1 isomers and enhanced (P < 0.05) total trans monounsaturated fatty acid and total conjugated linoleic acid content. Milk total n-3 polyunsaturated fatty acid (PUFA) content decreased with higher amounts of MS in the diet and n-6 PUFA concentration increased, leading to an elevated n-6 : n-3 PUFA ratio. Despite some beneficial changes associated with the replacement of GS with MS, the overall effects on milk FA composition would not be expected to substantially improve long-term human health. However, the role of forages on milk fat composition must also be balanced against the increases in total milk and protein yield on diets containing higher proportions of MS.     

Effects of conservation method on fatty acid composition of silage

This experiment was conducted to investigate effects of wilting and additives on the fatty acid (FA) composition of grass silage. The crop used was timothy (Phelum pratense L., cv. Grindstad), and the additives were Proens? (formic acid and propionic acid, 60–66 g/100 g and 25–30 g/100 g, respectively), the bacterial inoculant Siloferm® Plus (Pediococcus acidilactici and Lactobacillus plantarum) and water (control). The wilted material reached a dry matter (DM) content of 336 g/kg at the first cut and 350 g/kg at the second cut. Neither wilting nor the additives had any major effect on the FA proportions, with the only differences in the concentrations of C16:0 and C18:3. Silage treated with bacterial inoculant contained a higher proportion of C16:0 (P<0.05) than silage treated with acid, and a lower (P<0.05) concentration of C18:3 than silage treated with either of the other two additives. In the silages, there were lower (P<0.05) proportions of C16:0, C18:0, C18:1 and C18:3, and higher (P<0.05) proportions of C16:1, C18:2 and other identified FAs, than in the fresh material. A wilting process shorter than 24 h, to a DM content of 330–350 g/kg, did not have any effect on the proportions of FAs in P. pratense L., cv. Grindstad. Since the different additives and wilting strategies tested in this study did not affect the proportions of FAs in silage to a major extent, the results indicate that such a process offers a robust means to avoid losses of FAs that can occur during wilting, while retaining the positive effects of wilting, such as reduced losses of nutrients through effluents.     

Effect of chitosans on in vitro rumen digestion and fermentation of maize silage

The gas in vitro technique was used to study the effects of six types of chitosans, each having different molecular weights and acetylation degrees, on rumen microbial fermentation. In a first trial, a separate concentration of 750 mg/l of culture fluid for each of the six chitosans (CHI1, CHI2, CHI3, CHI4, CHI5, and CHI6) was incubated for 24 h in diluted ruminal fluid with maize silage as the substrate. The ionophore antibiotic monensin (MON) was used as a positive control, and a negative control with no chitosan (CTR) was also included. Each treatment was tested in triplicate for three different periods. At the end of the trial, samples were collected to determine volatile fatty acid (VFA) and ammonia N concentrations, and pH and gas production values were recorded. Methane concentration was estimated stoichiometrically. In vitro true organic matter digestibility (IVOMD) and partitioning factor (PF, mg OM truly degraded/ml gas produced) were also calculated. In a second trial, a separate concentration of 750 mg/l of each of the six chitosans was incubated for 144 h in diluted ruminal fluid with maize silage as the substrate, to study the effects of these compounds on fermentation kinetics.All six chitosans decreased the IVOMD and PF values. Chitosan inclusion did not affect the fermentation of the substrate's soluble fraction, but did reduce the fermentation kinetics of the insoluble but fermentable fraction. However, only CHI5 and CHI6 decreased total VFA concentration. CHI3 and CHI6 decreased the molar proportion of acetate and increased the molar proportion of propionate, thus increasing the propionate-to-acetate ratio. Chitosan inclusion did not affect molar proportions of butyrate. With the exception of CHI2, the molar proportion of branch-chained VFA was lowered by all of the chitosan treatments. Most of the treatments also decreased methane production, also with the exception of CHI2.In conclusion, chitosan extracts may enable the manipulation of rumen microbial fermentation, but further research is required to elucidate the effect of chitosans on ruminal fermentation parameters in commercial diets as well as the adaptability of rumen microflora to these additives.     

Insect feeding on different sorghum cultivars in relation to cyanide and phenolic acid content

Insect feeding on sorghum was studied in the field and in the laboratory. Quantitative differences in cyanogenesis and phenolic acid content were also measured on the same or simillar plants. High concentrations of cyanide were correlated with a reduction in feeding by grasshoppers and by first-instar larvae of Chilo partellus; high concentrations of phenolic acids were correlated with reduced feeding by various grasshoppers and by the planthopper Peregrinus maidis. Mythimna separata larvae, and adults and nymphs of Rhopalosiphum maidis, were apparently unaffected by these chemicals. Some sorghum cvs which were relatively unpalatable to grasshoppers and to P. maidis had low levels of the test chemicals, so that here some other factor or factors must be involved.     

Replacement of grass and maize silages with lucerne silage: effects on performance,milk fatty acid profile and digestibility in Holstein-Friesian dairy cows

In total, 20 multiparous Holstein-Friesian dairy cows received one of four diets in each of four periods of 28-day duration in a Latin square design to test the hypothesis that the inclusion of lucerne in the ration of high-yielding dairy cows would improve animal performance and milk fatty acid (FA) composition. All dietary treatments contained 0.55 : 0.45 forage to concentrates (dry matter (DM) basis), and within the forage component the proportion of lucerne (Medicago sativa), grass (Lolium perenne) and maize silage (Zea mays) was varied (DM basis): control (C)=0.4 : 0.6 grass : maize silage; L20=0.2 : 0.2 : 0.6 lucerne : grass : maize silage; L40=0.4 : 0.6 lucerne : maize silage; and L60=0.6 : 0.4 lucerne : maize silage. Diets were formulated to contain a similar CP and metabolisable protein content, with the reduction of soya bean meal and feed grade urea with increasing content of lucerne. Intake averaged 24.3 kg DM/day and was lowest in cows when fed L60 (P<0.01), but there was no effect of treatment on milk yield, milk fat or protein content, or live weight change, which averaged 40.9 kg/day, 41.0, 30.9 g/kg and 0.16 kg/day, respectively. Milk fat content of 18:2 c9 c12 and 18:3 c9 c12 c15 was increased (P<0.05) with increasing proportion of lucerne in the ration. Milk fat content of total polyunsaturated fatty acids was increased by 0.26 g/100 g in L60 compared with C. Plasma urea and β-hydroxybutyrate concentrations averaged 3.54 and 0.52 mmol/l, respectively, and were highest (P<0.001) in cows when fed L60 and lowest in C, but plasma glucose and total protein was not affected (P>0.05) by dietary treatment. Digestibility of DM, organic matter, CP and fibre decreased (P<0.01) with increasing content of lucerne in the diet, although fibre digestibility was similar in L40 and L60. It is concluded that first cut grass silage can be replaced with first cut lucerne silage without any detrimental effect on performance and an improvement in the milk FA profile, although intake and digestibility was lowest and plasma urea concentrations highest in cows when fed the highest level of inclusion of lucerne.     

Influence of host diet on the fatty acid composition and content of rumen protozoa in cattle

The fatty acid profiles and contents of protozoa from the rumen fluid of cattle varied according to the type of diet consumed by their host. Changing from a high-quality hay diet to a low-quality hay diet (DA) decreased the proportions of saturated acids and increased the proportions of the unsaturated acids 18:1 cis-9, 18:2 and 18:3 in neutral lipids (NL) and phospholipids (PL). Adding sucrose, urea and sulphur (SUS) to DA increased the proportions of branched chain acids in PL while addition of safflower oil increased polyunsaturated acids in PL and 18:1 trans-11 in NL. Diet did not alter the PL fatty acid content of protozoa but oil supplement of DA resulted in a 10-fold increase in the content of free fatty acids. The defaunating effect of oil supplement was partly reversed by SUS suggesting that factors other than the fatty acid content of cells are important in determining the toxicity of oil to rumen protozoa. The results indicate that the amounts of individual long-chain fatty acids taken up by rumen ciliates are largely determined by their concentrations in rumen digesta.     

Effects of dietary crude protein and tannic acid on rumen fermentation,rumen microbiota and nutrient digestion in beef cattle

The objectives of the trial were to study the effects of dietary crude protein (CP) and tannic acid (TA) on rumen fermentation, microbiota and nutrient digestion in beef cattle. Eight growing beef cattle (live weight 350 ± 25 kg) were allocated in a 2 × 2 crossover design using two levels of dietary CP [111 g/kg dry matter (DM) and 136 g/kg DM] and two levels of TA (0 and 16.9 g/kg DM) as experimental treatments. Each experimental period lasted 19 d, consisting of 14-d adaptation and 5-d sampling. The impacts of dietary CP and TA on ruminal microbiota were analysed using high-throughput sequencing of 16S rRNA gene. Results indicated that no interactions between dietary CP and TA were found on rumen fermentation and nutrient digestibility. Increasing dietary CP level from 111 to 136 g/kg DM increased the ruminal concentrations of ammonia nitrogen (NH₃-N) (p < 0.01) and improved the CP digestibility (p < 0.001). Adding TA at 16.9 g/kg DM inhibited rumen fermentation and decreased the digestibility of dietary CP (p < 0.001), DM (p < 0.05) and organic matter (p < 0.01). Increasing the dietary CP level or adding TA did not affect the relative abundances of the major bacteria Firmicutes and Proteobacteria at the phylum level and Prevotella_1 and Christensenellaceae_R-7_group at the genus level, even though adding TA increased the Shannon index of the ruminal bacterial community. TA was partly hydrolysed to pyrogallol, gallic acid and resorcinol in rumen fluid and the inhibitory effects of TA on rumen fermentation and nutrient digestibility could have been resulted from the TA metabolites including pyrogallol, gallic acid and resorcinol as well as the protein-binding effect.     

Nitrogen and fatty acid rumen metabolism in cattle offered high or low polyphenol oxidase red clover silage

Polyphenol oxidase (PPO) in red clover (RC) has been shown to reduce both lipolysis and proteolysis in silo and implicated (in vitro) in the rumen. However, all in vivo comparisons have compared RC with other forages, typically with lower levels of PPO, which brings in other confounding factors as to the cause for the greater protection of dietary nitrogen (N) and C18 polyunsaturated fatty acids (PUFA) on RC silage. This study compared two RC silages which when ensiled had contrasting PPO activities (RC+ and RC−) against a control of perennial ryegrass silage (PRG) to ascertain the effect of PPO activity on dietary N digestibility and PUFA biohydrogenation. Two studies were performed the first to investigate rumen and duodenal flow with six Hereford×Friesian steers, prepared with rumen and duodenal cannulae, and the second investigating whole tract N balance using six Holstein-Friesian non-lactating dairy cows. All diets were offered at a restricted level based on animal live weight with each experiment consisting of two 3×3 Latin squares using big bale silages ensiled in 2010 and 2011, respectively. For the first experiment digesta flow at the duodenum was estimated using a dual-phase marker system with ytterbium acetate and chromium ethylenediaminetetraacetic acid as particulate and liquid phase markers, respectively. Total N intake was higher on the RC silages in both experiments and higher on RC− than RC+. Rumen ammonia-N reflected intake with ammonia-N per unit of N intake lower on RC+ than RC−. Microbial N duodenal flow was comparable across all silage diets with non-microbial N higher on RC than the PRG with no difference between RC+ and RC−, even when reported on a N intake basis. C18 PUFA biohydrogenation was lower on RC silage diets than PRG but with no difference between RC+ and RC−. The N balance trial showed a greater retention of N on RC+ over RC−; however, this response is likely related to the difference in N intake over any PPO driven protection. The lack of difference between RC silages, despite contrasting levels of PPO, may reflect a similar level of protein-bound-phenol complexing determined in each RC silage. Previously this complexing has been associated with PPOs protection mechanism; however, this study has shown that protection is not related to total PPO activity.     

Influence of feeding whole sunflower seed and extruded linseed on production of dairy cows, rumen and plasma constituents, and fatty acid composition of milk

Holstein cows were fed total mixed rations (TMR) supplemented with protected palm fat (PPF), whole sunflower seed (WSS) or extruded linseed (ELS) for 100 days. Percentage of dietary crude fat was 5.3, 5.1 and 5.1, respectively. Diet had no (p > 0.05) effect on feed intake, milk yield or milk protein content. Percentage of milk fat and yield of fat--corrected milk were significantly increased when diets were supplemented with WSS and ELS. Feeding PPF resulted in the lowest (p < 0.05) ruminal concentration of volatile fatty acids. No significant dietary effect on plasma characteristics was observed. Concentration of polyunsaturated fatty acids (PUFA) was higher (p < 0.05), and PUFA n-6/n-3 ratio lower (p < 0.05), in the milk fat from cows fed ELS compared to WSS. Supplementation of TMR with oilseeds compared to PPF increased the content of CLA in milk fat (p < 0.005) and decreased its atherogenicity, primarily due to a significant reduction of palmitic acid concentration. Both oilseeds significantly improved the spreadability index of manufactured butter. ELS, but not WSS, increased the susceptibility of milk fat to oxidation (p < 0.05). It can be concluded that feeding of oilseeds to dairy cows improved nutritional quality of milk fat, with supplementation with ELS producing an even more desirable milk fatty acid profile than WSS supplementation.     

Influence of feeding whole sunflower seed and extruded linseed on production of dairy cows,rumen and plasma constituents,and fatty acid composition of milk

Holstein cows were fed total mixed rations (TMR) supplemented with protected palm fat (PPF), whole sunflower seed (WSS) or extruded linseed (ELS) for 100 days. Percentage of dietary crude fat was 5.3, 5.1 and 5.1, respectively. Diet had no (p > 0.05) effect on feed intake, milk yield or milk protein content. Percentage of milk fat and yield of fat – corrected milk were significantly increased when diets were supplemented with WSS and ELS. Feeding PPF resulted in the lowest (p < 0.05) ruminal concentration of volatile fatty acids. No significant dietary effect on plasma characteristics was observed. Concentration of polyunsaturated fatty acids (PUFA) was higher (p < 0.05), and PUFA n-6/n-3 ratio lower (p < 0.05), in the milk fat from cows fed ELS compared to WSS. Supplementation of TMR with oilseeds compared to PPF increased the content of CLA in milk fat (p < 0.005) and decreased its atherogenicity, primarily due to a significant reduction of palmitic acid concentration. Both oilseeds significantly improved the spreadability index of manufactured butter. ELS, but not WSS, increased the susceptibility of milk fat to oxidation (p < 0.05). It can be concluded that feeding of oilseeds to dairy cows improved nutritional quality of milk fat, with supplementation with ELS producing an even more desirable milk fatty acid profile than WSS supplementation.     

The fatty acid composition of milk fat as influenced by feeding oilseeds

The fatty acid composition of bovine milk fat can be substantially altered by feeding lipid sources which alter the fatty acid profile of lipid entering the intestine from the rumen. As long-chain fatty acids of dietary origin can be incorporated directly into milk fat the opportunity exists to alter the ratio of short and long-chain fatty acids as well as the degree of saturation of milk fat. In practice our ability to alter the fatty acid profile of milk fat is limited not by the synthetic capacity of the mammary gland, but rather by the challenge of achieving effective protection of unsaturated dietary fatty acids from biohydrogenation in the rumen, as well as keeping the level of polyunsaturated fatty acids within the range where the organoleptic quality and shelf-life of milk and dairy products are not compromised. The fatty acid composition of oilseeds such as canola are considered desirable from a human health perspective and thus their inclusion in the diet of dairy cattle as a means of achieving a more desirable fatty acid profile in milk fat may enhance the nutritive quality of milk.     

Effect of maize silage to grass silage ratio and feed particle size on protein synthesis and amino acid profile in different microbial fractions in a semi-continuous rumen simulation

The objective was to investigate the effect of variation in forage source and feed particle size of a diet, including interactions, on the amount and the composition of microbial crude protein (CP) in a semi-continuous culture system (Rusitec). Different microbial CP fractions were compared. Five diets with mean forage proportion of 0.88 and different maize silage to grass silage ratios (100 : 0, 79 : 21, 52 : 48, 24 : 76 and 0 : 100) were used. Diets were ground through sieves with a pore size of either 1 or 4 mm, matching the particle size of fine (F) and coarse (C), respectively. Diets were characterised by increasing concentrations of CP and fibre fractions, and decreasing concentrations of starch with ascending inclusion rates of grass silage. Microbial mass was isolated from feed residues after incubation from the liquid phase of the fermenter and from the liquid effluent. The amount of synthesised microbial CP was determined on the basis of 15N balance. It increased quite linearly by the stepwise replacement of maize silage by grass silage, and was higher in C treatments compared to F treatments. Efficiency of microbial CP synthesis (EMPS) was improved from 29 to 43 mg microbial N/g degraded organic matter (OM) by increasing the proportion of grass silage in the diet, but was unaffected by particle size. The N content as well as the profiles of amino acids of the three microbial fractions was affected by diet composition and particle size. The ratio of solid- to liquid-associated microbes was affected by diet composition and feed particle size. The amount and EMPS seemed to be improved by degradation of OM from grass silage and an increasing availability of N. Moreover, the results of this study indicated a shift in the composition of the microbial community caused by variation in forage composition and feed particle size.     

Fatty acid mobilization and comparison to milk fatty acid content in northern elephant seals

A fundamental feature of the life history of true seals, bears and baleen whales is lactation while fasting. This study examined the mobilization of fatty acids from blubber and their subsequent partitioning into maternal metabolism and milk production in northern elephant seals (Mirounga angustirostris). The fatty acid composition of blubber and milk was measured in both early and late lactation. Proportions of fatty acids in milk and blubber were found to display a high degree of similarity both early and late in lactation. Seals mobilized an enormous amount of lipid (~66 kg in 17 days), but thermoregulatory fatty acids, those that remain fluid at low temperatures, were relatively conserved in the outer blubber layer. Despite the stratification, the pattern of mobilization of specific fatty acids conforms to biochemical predictions. Long chain (>20C) monounsaturated fatty acids (MUFAs) were the least mobilized from blubber and the only class of fatty acids that showed a proportional increase in milk in late lactation. Polyunsaturated fatty acids (PUFAs) and saturated fatty acids (SFAs) were more mobilized from the blubber, but neither proportion increased in milk at late lactation. These data suggest that of the long chain MUFA mobilized, the majority is directed to milk synthesis. The mother may preferentially use PUFA and SFA for her own metabolism, decreasing the availability for deposition into milk. The potential impacts of milk fatty acid delivery on pup diving development and thermoregulation are exciting avenues for exploration.     

Effects of feeding level and protein content of milk replacer on the performance of dairy herd replacements

It has been suggested that United Kingdom recommendations for feeding the neonatal calf (500 g milk replacer (MR)/day; 200-230 g CP/kg milk powder) are inadequate to sustain optimal growth rates in early life. The current study was undertaken with 153 high genetic merit, male and female Holstein-Friesian calves (PIN2000 = ￡48) born between September and March, with heifers reared and bred to calve at 24 months of age. Calves were allocated to one of four pre-weaning dietary treatments arranged in a 2 MR feeding level (5 v. 10 l/day) × 2 MR protein content (210 v. 270 g CP/kg dry matter (DM)) factorial design. MR was reconstituted at a rate of 120 g/l of water, throughout, and was offered via computerised automated milk feeders. Calves were introduced to pre-weaning diets at 5 days of age and weaned at day 56. During the first 56 days of life, calves offered 10 l MR/day had significantly higher liveweight gains (P < 0.001) than calves fed 5 l MR/day. No significant differences in liveweight gain were found between calves fed 210 g CP/kg DM MR and those fed 270 g CP/kg DM MR from birth to day 56. Differences in live weight and body size due to feeding level disappeared by day 90. Neither MR feeding level nor MR CP content affected age at first service or age at successful service, and with no milk production effects, the results indicate no post-weaning benefits of increased nutrition during the milk-feeding period in dairy heifers.     

Effects of medium-chain fatty acid supplementation on performance and rumen fermentation of lactating Holstein dairy cows

Medium-chain fatty acids (MCFAs) have antimicrobial properties and cause negative or positive effects on animal performance depending on its dosage. We hypothesized that MCFA supplementation at a lower dose (i.e., 0.05–0.2% of dietary DM) would increase rumen pH and milk production without decreasing nutrient digestibility which is typically observed with the higher inclusion rates (i.e., >1% of dietary DM). The objective of this study was to evaluate the effects of MCFA supplementation at a lower dose on productivity, plasma energy metabolite concentrations, apparent total tract nutrient digestibility, rumen fermentation, and rumen microbial profile of lactating dairy cows. Thirty (n = 8 primiparous, n = 22 multiparous) Holstein cows in mid-lactation (637 ± 68.5 kg of initial BW, 98.5 ± 27.4 d in milk; mean ± standard deviation) were used in a crossover design with two 28-d periods. The MCFA supplement, consisted of 25% MCFA (containing 32% C8:0, 21% C10:0, 47% C12:0 on DM basis) and 75% carrier ingredients, was fed at 0.25% of dietary DM replacing dry ground corn in control (CON). Total inclusion of MCFA was 0.063% of dietary DM. No differences were observed in DM intake, apparent total tract nutrient digestibility and BW change between MCFA and CON. Milk and milk component yields did not differ between treatment groups. The MCFA supplementation tended to have higher minimum rumen pH (5.66 vs. 5.54), and decreased daily fluctuation range of rumen pH (1.17 vs. 1.40) compared to CON. However, the duration of acidosis (pH < 5.8, min/d) did not differ between treatment groups and ruminal total volatile fatty acid concentration and its profile did not differ between treatment groups. For rumen microbiota, the Chao1 index of bacterial community tended to be lower (10.9 vs. 11.6) whereas the Shannon index did not differ (0.91 vs. 0.93) in MCFA compared to CON, and both indices did not differ for archaeal and protozoan communities between treatment groups. The relative abundance of Methanobrevibacter gottschalkii increased when supplemented with MCFA (5.14 vs. 4.92%). These results suggest that supplementation of MCFA at 0.063% dietary DM may not affect overall animal performance or total tract nutrient digestibility, but decrease the daily range of pH and the bacterial richness in the rumen.     

Effects of natural and commercial diets on the fatty acid content of European grayling

Eicosaenoic acid (20:1ω9) and docosaenoic acid (22:1ω11) levels were about 10 and 100 times higher in food pellets fed to cultured grayling than in the insect larvae on which wild grayling fed. Among the PUFA, docosahexaenoic acid (DHA) was very high in the pellets, resulting in an unnaturally elevated, and probably unbalanced, ω3/ω6 ratio of 7–13 in the cultured fish whereas the same ratio varied only from 4 to 6 in the wild fish. Despite very low DHA levels in the native food, wild grayling muscle tissue contained relatively high amounts of DHA. DHA is probably not essential in the diet of grayling.     

Characterization of gut microbiota and short-chain fatty acid in breastfed infants with or without breast milk jaundice

This study aims to investigate the gut microbiota and metabolites in breastfed infants with breast milk jaundice (BMJ) using gut microbiome–metabolomics. Breastfed newborns diagnosed with BMJ and those without BMJ (control group) were enrolled. Faecal samples were collected from the participants and subjected to high-throughput sequencing of the 16s rDNA V3 and V4 regions of the gut flora and metabolomics of short-chain fatty acids (SCFAs). Proteobacteria, Fimicutes and Actinobacteria were the main bacteria at the phylum level. Eshcerichia-Shigella and Enterobacteriacea were the main bacteria at the genus level. The difference between the two groups was compared. Compared to the control group, the amount of Streptococcus was significantly increased while the amount of Enterococcus was significantly decreased in the faeces from infants with BMJ. Functional prediction analysis of 16S found that biosynthesis of penicillin and cephalosporin significantly increased in the BMJ group. Gas chromatography–mass spectrometry detection of SCFAs revealed that levels of acetic acid and propionic acid were significantly lower in the BMJ group than in the control group. The reduced levels of acetic acid and propionic acid may be related to the increase in Streptococcus and decrease in Enterococcus, both of which may contribute to BMJ.