不同实验室间测定单胃动物饲料样品总能的一致性及可加性研究

Consistency and Additivity in Gross Energy of Monogastric Animal Feed Samples between Different Laboratories

本研究旨在比较不同实验室间测定饲料总能(GE)的变异程度及可加性,为饲料有效能值的准确测定提供参考。试验分为2个部分。内容一:考察饲料的上样量对氧弹热量计测定GE值的影响,采用单因素完全随机设计,6个饲料原料分别为玉米、豆粕、小麦麸、大豆油、菜籽粕、棉籽粕,10个饲粮上样量共计5个处理,分别设为0.2、0.4、0.6、0.8和1.0 g,每个处理4个重复。内容二:考察实验室间测定饲粮GE的变异及可加性。采用单因素完全随机设计,共6个实验室,每个实验室对各样品测定4个重复。结果表明:6个饲料原料及10个饲粮GE的测定值均随上样量的增加而呈显著线性和二次曲线增加(P<0.05)。0.2 g上样量的GE测定值均显著低于0.4~1.0 g上样量相应的GE测定值(P<0.05)。在0.6~1.0 g上样量间,样品GE测定值相对差异较小。在6个实验室对比中,96个测定数据中的91个的|Z|<2。h统计量表明,实验室2的测定结果偏低,而实验室5的测定结果偏高。k统计量表明,实验室5和6的测定变异相对较大。总体上,GE测定的重复性变异系数、实验室间变异系数及再现性变异系数不超过0.64%。配对t检验表明,实验室1、3、5和6对10个饲粮GE的计算值与测定值差异不显著。而实验室2和4对10个饲粮GE的计算值与测定值差异显著(P<0.05)。6个实验室内饲粮的GE测定值对计算值的线性回归模型与Y=X重叠。由此表明,氧弹热量计测定GE的上样量应在0.6 g以上。6个实验室间测定饲粮GE值呈现较好的一致性(94.8%的满意度),实验室间的变异系数均较低。4个实验室测定饲粮GE的可加性高,2个实验室的可加性稍低。

The objective of this study was to compare the variation and additivity of gross energy(GE)measured in different laboratories,which will provide a reference for accurately determining effective energy value of feed.This experiment consisted of two parts.Part 1 was to investigate the effects of the feed sample volumes on the GE value determined by the oxygen bomb calorimeter.A single factor completely random design for sample volumes at 0.2,0.4,0.6,0.8 and 1.0 g was adopted for each of 6 feed ingredients(corn,soybean meal,wheat bran,soybean oil,rapeseed meal and cottonseed meal)and 10 diets.Each treatment contained 4 replicates.Part 2 was to investigate the variation and additivity of GE in feed determined in different laboratories.A single factor completely random design for 6 laborotaries was adopted for each feed.The GE of each sample was measured for 4 replicates in each laboratory.The results showed that the measured GE of 6 feed ingredients and 10 diets increased linearly and quadraticly with the increment of sample volumes(P<0.05).The measured GE values at sample volume of 0.2 g were lower than the GE values measured at sample volume of 0.4 to 1.0 g(P<0.05).Relatively little difference was observed in GE measured at sample volumes of 0.6 to 1.0 g.In the comparison among 6 laboratories,the absoute Z scores of 91 out of 96 measured GE were less than 2.The h statistics showed that the measured GE in laboratory 2 was lower,whereas the measured GE in laboratory 5 was higher than other laboratories.The k statistics showed that the variation of GE measured in laboratory 5 and 6 was relatively greater.In total,the coefficient of variation for repeatability,inter-laboratory and reproducibility of GE didn’t exceed 0.64%.The paired t test showed that no significant differences between the calculated and measured GE of 10 diets were observed in each of laboratory 1,3,5 and 6.Whereas,significant differences between the calculated and measured GE of 10 diets were observed in each of laboratory 2 and 4(P<0.05).The linear regression model of measured on calculated GE in each of 6 laboratories all overlaped with Y=X.This indicates that the loading volumes of sample for GE measurement using the oxygen bomb calorimeter should be more than 0.6 g.The GE value of feed measured in 6 laboratories show good consistency(94.8%satisfaction),and low coefficient of variation is observed in GE measured among laboratories.Higher and high additivity of dietary GE are present in laboratory 4 and 2,respectively.