黄梁木环割促萌正交试验及混合效应模型分析  

Orthogonal test and mixed effect model analysis for the germination promotion of Neolamarckia cadamba by girdling

作　　者：阙青敏 谭锦豪 张登 欧阳昆唏 陈晓阳 李华强 QUE Qing-min;TAN Jin-hao;ZHANG Deng;OUYANG Kun-xi;CHEN Xiao-yang;LI Hua-qiang(College of Forestry and Landscape Architecture,South China Agricultural University/Guangdong Key Laboratory for Innovative Development and Utilization of Forest Plant Germplasm,Guangzhou 510642,China;State-run Leizhou Forestry Bureau,Zhanjing,Guangdong 524348,China)

机构地区：[1]华南农业大学林学与风景园林学院/广东省森林植物种质创新与利用重点实验室,广州510642 [2]国营雷州林业局,广东湛江524348

出　　处：《南方农业学报》2020年第3期617-622,共6页Journal of Southern Agriculture

基　　金：中央财政林业科技推广示范项目(〔2018〕GDTK-08);广东省科技计划项目(2017B020201008);广东省自然科学基金项目(2015A030313419);广州市科技计划项目(201607020024)。

摘　　要：【目的】开展黄梁木环割促萌正交试验并进行混合效应模型分析,为培育黄梁木高幼化程度无性繁殖材料及黄梁木优良种质资源保存和推广应用打下基础。【方法】以黄梁木成年大树为试验材料,设环割高度和宽度2个因素,环割宽度分别为2、3和4 cm及环割高度分别为20、30和40 cm 3个水平的正交试验,共9个处理(处理1~9);环割后每隔1周调查并记录各处理的不定芽发芽数;利用双因素方差分析和线性混合效应模型,分析筛选黄梁木环割促萌最佳条件组合。【结果】处理1的平均发芽数最多,为6.38个,明显多于处理2、处理4、处理8和处理9,显著多于处理3、处理5、处理6和处理7(P<0.05,下同)。双因素方差分析结果显示,环割高度和环割宽度对发芽数影响不显著(P>0.05,下同),但二者的交互作用对发芽数具有极显著影响(P<0.01);线性混合效应模型分析结果表明,环割高度、环割高度×初始胸径和环割宽度×初始胸径对发芽数有显著影响,环割宽度、初始胸径和环割宽度×环割高度对发芽数无显著影响。【结论】经双因素方差分析和线性混合效应模型分析,黄梁木环割促萌的最佳条件为环割高度20 cm、环割宽度2~3 cm。【Objective】In order to lay a foundation for the preservation and promotion of elite germplasm on Neolamarckia cadamba,and to provide a reference for the asexual propagation material with high degree of rejuvenation,an orthogonal experiment on girdling promoting germination of N.cadamba and mixed effect model analysis were carried out.【Method】The experiment was carried out on mature N.cadamba,the orthogonal test included two factors,girdling height(GD)and girdling width(GH),and three levels for each factor(GH were 2,3 and 4 cm,GD were 20,30 and 40 cm),a total of nine treatments were processed(treatment 1-9).The number of adventitious buds of each treatment was investigated and recorded weekly after girdling.The two-factor analysis of variance and the linear mixed effect model were used to analyze the statistics and select the optimal combination of conditions.【Result】The average buds number in treatment 1 was the largest,with an average number of 6.38,more than those in treatment 2,treatment 4,treatment 8 and treatment 9,significantly more than those in treatment 3,treatment 5,treatment 6 and treatment 7(P<0.05,the same below).The results of two-way analysis of variance showed that the GH and GD had no significant effect on the number of germination bud(P>0.05,the same below),but the interaction between GH and GD had extremely significant effect on the germination number(P<0.01).The linear mixed-effects model analysis showed that the GH,GH×initial diameter at breast height(DBH),GD×DBH had significant effect on germination number.The GD,DBH,GH×GD had no significant effect on the number of germination.【Conclusion】The optimum conditions for the girdling of N.cadamba are 20 cm in GH and 2-3 cm in GD according to two-way analysis of variance and linear mixed-effects model analysis.

关 键 词：黄梁木 环割 正交试验设计 线性混合效应模型 

分 类 号：S722.3[农业科学—林木遗传育种]