Jianming HAO, Gang LIU, Quanhong OU, Xiangping ZHOU

School of Physics and Electronic Information, Yunnan Normal University, Kunming 650500, China

Zea mays (Poaceae:Maydeae)is annual cereal plant. It originated from North, Central and South America. Corn plants are high,and have strong and straight stems.Corn is composed by pericarp, endosperm, cotyledons, germ, hypocotyl and radicle. It is one of the major food crops for human. Corn is known as longevity food,and it is rich in proteins,fat, vitamins, trace elements, cellulose and polysaccharides. Since Yunnan Province has a unique geographical location and natural climatic conditions, corn can be planted throughout the year in Yunnan Province. Spring corn is mainly distributed in single cropping areas in northeastern and northwestern Yunnan,summer corn is distributed in double cropping areas in central Yunnan and mountain areas in southern Yunnan, and autumn and winter corn is distributed in riverside valley areas and multiple cropping areas in southern Yunnan.

Corn can be divided by many different ways.In trade,corn is only divided into flour corn and colloid corn. According to corn color, corn can be divided into yellow corn, white corn and yellow-white corn. According to grain morphology and structure,corn can be divided into flint type, dent type, flour type,sweet type,sweet flour type,pop type, waxy type, pod corn and halfdent type. According to growth period,corn can be divided into early, middle and late mature varieties.According to grain composition and use, corn can be classified into special and common corn. The special corn can further divided into high lysine corn, waxy corn, sweet corn[1], pop corn and high oil corn.

The seed coat color-based classification is the most simple and intuitive classification. However, the nutrients and their contents are not involved.The analysis methods of corn seed ingredients include near-infrared analysis, NMR, chemical analysis and infrared analysis. These methods all have their own advantages and disadvantages. Fourier transform infrared spectroscopy method can quickly and accurately detect the main components in corn, and it is characterized by rapidness, simplicity, non destruction and easy operation. Currently,FTIR technology has been widely used in biological,chemical,food,agricultural,forestry and medical researches. In recent years, there have been also many reports on application of FTIR in identification and classification of plants,herbs,nuts,etc[2-8].Sun et al.[9-10]studied ganoderma by using FTIR.Liu et al.[11-12]identified wild edible mushrooms by FTIR spectroscopy. Zhou et al.[13-14]studied and classified Boletus speciosus and black fungus of different origins by using FTIR technology combined with cluster analysis. Li et al.[15]studied the ingredients in corn seeds by using IR and Raman spectroscopy. Yan et al.[16]studied the characteristics of FTIR spectra of corn endosperm and germ. Shi et al.[17]investigated the grain quality of five kinds of hybrid corn by using FTIR.

Therefore, this study compared and clustered the corn germs and endosperms of three different colors in Yunnan Province by using FTIR combined with hierarchical cluster analysis.

Material and Methods

Instruments and equipment and test conditions

The used Fourier transform infrared spectroscopy (FTIR Frontier)was produced by the Perkin Elmer Company, US. Its scan times, resolution and scanning range were 16, 4 cm-1and 4 000-400 cm-1,respectively.All the spectra were collected by using Spectrum software version 10.3.4 and processed by using Omnic 8.0. The cluster analysis was performed by using IBM SPSS Statistics 20.

Sample preparation

A total of 52 samples, from three kinds of corns, were prepared. The corns were all provided by the Yunnan Academy of Agricultural Sciences. Before the determination, the samples were all washed with water and then dried in the sun. During the detection,the 52 corn samples were first dried in oven (80 ℃). Subsequently, certain amount of germs and endosperms were ground in two agate mortars.After mixed with KBr according to proportion of 1:100, the samples were detected with FTIR. Before the formal detection started, the background spectrum had already been deducted by using Spectrum software version 10.3.4. All the spectra of samples were pre-processed with nine-point smoothing, baseline correction and normalization.

Results and Analysis

Characteristics of infrared spectra of three types of corns

The original infrared spectra of the three kinds of corn germs were shown in Fig.1.In Fig.1,a represented the averaged infrared spectrum of bnby(germ of white waxy corn); b represented averaged infrared spectrum of htby (germ of yellow sweet and crisp corn); c represented the averaged infrared spectrum of lsby (germ of double-color sweet and crisp corn). Fig.2 showed the original infrared spectra of the three kinds of corn endosperms.In Fig.2, d represented the averaged infrared spectrum of bnb (endosperm of white waxy corn); e represented the averaged infrared spectrum of htb(endosperm of yellow sweet and crisp corn); f represented the averaged infrared spectrum of lsb (endosperm of double-color sweet and crisp corn).

As shown in Fig.1, the peaks of corn germs were mainly shown at the wavelengths of 3 367, 2 928, 1 653,1 671 and 1 246 cm-1, and the peaks of corn endosperms were mainly shown at the wavelengths of 3 410,2 930,1 571,1 420 and 1 246 cm-1.At the wavelength of 3 367 cm-1, the peaks were mainly generated by the stretching vibration of hydroxyl and amidogen from polysaccharides and proteins. At the wavelengths of 2 928 and 2 930 cm-1,the peaks were mainly generated by the asymmetric stretching vibration and symmetric stretching vibration of methyl and methylene from lipids, polysaccharides and proteins.At the wavelengths of 1 635 and 1 571 cm-1,the peaks were generated by the stretching vibration of C=O and the bending vibration of N-H from amid I and II bands.The wavelength range of 1 500-1 200 cm-1was the absorption area of mixed vibration of proteins,fatty acids and polysaccharides. The highest peaks of germs were shown around the wavelength of 2 928 cm-1,which was the absorption area of lipids; while the highest peaks of endosperms were shown around the wavelength of 1 000 cm-1, which was the absorption area of carbohydrates.

Differences in infrared spectra of three kinds of corn germs and endosperms

The comparison of the spectra of the 52 corn germ and endosperm samples showed that the originalspectra were overall similar, indicating the similar major chemical components.However,there were still certain differences in peak number, peak position and peak intensity among the three colors of corns within the wavelength rage of 1 800-700 cm-1,indicating the fingerprinting and characteristic features (Fig.1). At the wavelength of 3 367 cm-1,the peak intensities of corn endosperms were higher than those of corn germs, indicating higher contents of polysaccharides in corn endosperms. At the wavelength of 1 000 cm-1, the peak intensities of corn endosperms were also higher than those of corn germs, indicating higher contents of carbohydrates in corn endosperms. The differences in spectrum among the three colors of corns,shown in Fig.1,were mainly caused by the differences in chemical components. However, in overall, the differences were not very obvious. In order to better reflect the differences in spectrum within the wavelength range of 1 000-1 800 cm-1,the first derivative(Fig.2) and second derivative (Fig.3)spectra of corn germs and endosperms of three different colors were obtained.

As shown in Fig.2(left),there were high-intensity peaks shown at the wavelength of 1 748 cm-1, and they were generated by the vibration absorption of carbonyl of lipids in corn germs.However,in the spectra of corn endosperms, there were no obvious peaks shown at the wavelength of 1 748 cm-1. In the absorption area of amide I band, the absorption peaks of corn germs and endosperms were all not obvious. However, at the wavelength of 1 000 cm-1,the peak intensities of corn endosperms were significantly higher than those of corn germs, indicating higher contents of carbohydrates in corn endosperms.

In the second derivative spectra of corn germs of three colors (Fig.3),high-intensity negative peaks were shown at the wavelength of 1 748 cm-1.While in the second derivative spectra of corn endosperm of three colors, although there were also negative peaks shown in spectra of e and f, their intensities were significantly lower than those negative peaks of corn germs;there were no obvious negative peaks shown in the spectrum of d.It was indicated that the lipids contents in corn endosperms were lower than those in corn germs. At the wavelength of 1 000 cm-1,the peak intensities of corn endosperms were significantly higher than those of corn germs, indicating higher carbohydrates contents in corn endosperms.

Correlation analysis and hierarchical cluster analysis of three kinds of corn

Correlation analysis can be performed by calculating the correlation coefficients among the samples.It can be used to measure the degree of linear correlation between variables.Thus the degree of similarity between two samples can also be obtained.Within the wavelength range of 700-1 800 cm-1,the correlation coefficients among the second derivative spectra of corn germs and endosperms of three colors were calculated according to the following equation:

Wherein, r represented correlation coefficient; xiand yirepresented the absorbances of x and y spectra at the frequency of i; x and y represented the average absorbances of x and y spectra.

The correlation analysis showed that the correlation coefficients of the averaged first derivative and second derivative spectra among the corn germs and among the corn endosperms of the same color were all more than 0.908. It was indicated that the chemical components or relativecontents of chemical components were the same among the corn germs and among the corn endosperms of the same color. However, there were large differences in correlation coefficients of the first derivative and second derivative spectra between corn germs and endosperms of the same color or different color.For example,the correlation coefficients of the first derivative spectra between corn germs and endosperms of white waxy corn ranged from 0.364 to 0.989, the correlation coefficients of the first derivative spectra between corn germs and endosperms of yellow sweet and crisp corn ranged from 0.364 to 0.988, and the correlation coefficients of the first derivative spectra between corn germs and endosperms of double-color sweet and crisp corn ranged from 0.435 to 0.996; the correlation coefficients of the second derivative spectra among corn germs, between corn germs and endosperms or among corn endosperms of white waxy corn ranged from -0.298 to 0.994, the correlation coefficients of the second derivative spectra among corn germs, between corn germs and endosperms or among corn endosperms of yellow sweet and crisp corn ranged from -0.282 to 0.987,and the correlation coefficients of the second derivative spectra among corn germs, between corn germs and endosperms or among corn endosperms of double-color sweet and crisp corn ranged from -0.268 to 0.990. In overall, the correlation coefficients differed greatly among corn germs, among corn endosperms and between corn germs and endosperms of the same color and different color.It was indicated that the chemical components or relative contents of chemical components had changed to some extent.

After the comparison of correlation coefficients of first derivative and second derivative spectra among corn germs, among corn endosperms and between corn germs and endosperms of three different colors,cluster analysis was further performed for the second derivative spectra. Based on the observation on multiple indicators of the same-batch samples, hierarchical cluster analysis can find some statistics that can measure the degree of similarity between samples by using quantitative mathematical methods.The found statistics will be further used as the basis for classification. The basic idea of HCA is to cluster some samples with high similarities into one group and to cluster the other samples with high similarities into another group. The samples with higher similarities will be clustered into a small group, while the samples with lower similarities will be clustered into a large group.This cycle will be ended until all the samples are clustered. This chemometric method has great practical value,so it has been widely used in recent years.Hao et al.[18]classified the Caprifoliaceae plants by using cluster analysis. Hong et al.[19]identified dodder by using cluster analysis.Li et al.[20]studied infected leaves of broad bean by using infrared spectroscopy and cluster analysis.

The absorbance matrices were constructed with the absorbances of the second derivative spectra of all the samples within the wavelength range of 700-1 800 cm-1.Then the Euclidean distance-based cluster analysis of different-color corns was performed by using the Ward's method. The results showed that, within the wavelength range, the second derivative spectra were better clustered according to different origins. As shown in Fig.4, the corn germs and endosperms of the corns of three different colors could be better clustered according to different types, which was basically consistent with actual situation.There was certain correlation between type and classification. In the dendrogram, the samples could be divided into different groups according to different classification levels. As shown in Fig.4, the 52 corn germ and endosperm samples were divided into six large groups,which was consistent with the classification results according different colors and different parts. At different classification levels, the samples were classified into different groups. However, as shown in the dendrogram, if the classification groups were assigned as six, the samples could be better clustered according to different colors and different parts. The correct rate of classification could reach 96.1%,indicating relatively satisfactory classification results.

Conclusions and Discussion

The results showed that the peak shapes and positions of original infrared spectra of corn germs and endosperms were relatively similar among the three types. It was indicated that the chemical components,mainly including polysaccharides, proteins, lipids and carbohydrates, were the same among the three colors of corns. However, in the wavelength range of 700-1 800 cm-1, there were certain differences in peak number,peak position and peak shape.The differences in peak intensity at the wavelength of 3 367 cm-1indicated that the polysaccharides contents in corn endosperms were higher thanthose in the corn germs; the differences in peak intensity at the wavelength of 2 928 cm-1indicated that the fat contents in corn endosperms were lower than those in the corn germs;the differences in peak intensity at the wavelength of 1 000 cm-1indicated that the carbohydrates contents in corn endosperms were higher than those in the corn germs. In terms of negative peaks, the differences at 1 748 cm-1indicated that the fat content in corn germs were higher than those in corn endosperms; the differences at 1 000 cm-1indicated that the carbohydrates contents in corn germs were lower than those in corn endosperms. The cluster analysis and correlation analysis of three colors of corns were performed within the wavelength range of 700-1 800 cm-1. The results showed that the correlation coefficients of averaged first derivative and section derivative spectra among corn germs and among corn endosperms of the same colors were all more than 0.908, indicating similar chemical components or similar relative contents of chemical components in corn germs and endosperms among different colors of corns. However,there were large differences in correlation coefficients of first derivative and second derivative spectra between corn germs and endosperms of the same color,indicating certain changes in chemical components and relative contents of chemical components in corns.The hierarchical cluster analysis was performed for the 52 corn germ and endosperm samples by using Ward’s method. The results showed the same-color corn could be divided into corn germs and corn endosperms,which was consistent with the actual classification results according to different colors and different parts. The correct rate of classification could reach 96.1%.

In this study, the FTIR analysis results show that the basic components in corn germs and endosperms are the same among different colors.However, there are certain differences in contents of the components. So there may be certain difference in taste. The samples are limited, indicating certain reference value of these study results.In order to make general conclusions, large number of samples and further study are required.In addition, there are differences in taste and texture among different colors of corns,which may be caused by different proportions of basic components in white waxy corn, yellow sweet and crisp corn and double-color sweet and crisp corn.The quantitative analysis of FTIR spectra can also reveal the proportions of basic components in each of the samples,which is still needed to be studied further.

[1]YANG XG (楊孝臣), LI YY (李妍妍),FENG G (豐光).Classification and biological characteristics and cultivation techniques of wheat corn(甜玉米的分類和生物學(xué)特征及其栽培技術(shù))[J].Modern Agricultural Science and Technology(現(xiàn)代農(nóng)業(yè)科技),2008(19):236,238.

[2]DOGAN A, SIYAKUS G, SEVENRCAN F. FTIR spectroscopic characterization of irradiated hazelnut (Corylus avellana L.) [J]. Food Chemistry, 2007, 100:1106-1114.

[3]LUO BR(羅庇榮),LIU G(劉剛),SHI YM(時有明), et al. Discrimination of Ericaceae using FTIR spectroscopy and principal component analysis (FTIR 結(jié)合主成分分析對杜鵑花植物鑒別分類研究)[J].Infrared Technology (紅外技術(shù)),2009,31(1):39-43.

[4]ZUO K(左康), LI DT(李丹婷), GUO SL(郭水良), et al. FTIR spectra-principal component analysis of roots of Polygonum cuspidatum from different areas(不同產(chǎn)地虎杖根FTIR 主成分分析研究)[J].Spectroscopy and Spectral Analysis(光譜學(xué)與光譜分析), 2007, 27(10):1989-1992.

[5]BONNIN S, BESSON F, GELHAUSEN M. A FTIR spectroscopy evidence of the interactions between wheat germ agglutinin and N-acetyl glucosamine residues [J]. FEBS Letters, 1999, 456:361-364.

[6]GUIBETA F,AMIELA C,CADOTB P, et al. Discrimination and classification of enterococci by Fourier transform infrared (FTIR) spectroscopy [J]. Vibrational Spectroscopy, 2003, 33: 133 -142.

[7]NGOTHI NA, KIRSCHNER C, NAUMANN D. Characterization and identification of microorganisms by FTIR microspectrometry[J]. Journal of Molecular Structure,2003,661-662:371-380.

[8]GORGULU ST, DOGAN M, SEVERCAN F.The characterization and differentiation of higher plants by Fourier transform infrared spectroscopy[J].Applied Spectroscopy,2007,61 (3): 300-308.

[9]SUN SQ (孫素琴), DU DG (杜德國),LIANG XY ( 梁曦云), et al. A rapid method for distinguishing the different Ganoderma lucidum products by Fourier transform infrared spectroscopy (36種靈芝產(chǎn)品傅里葉變換紅外光譜快速鑒別研究)[J].Chinese Journal of Analytical Chemistry (分析化學(xué)研究簡報),2001,29(3):309-312.

[10]SUN SQ(孫素琴),LIANG XY(梁曦云),YANG XR(楊顯榮).A rapid method for detecting Ganoderma lucidum of six colors by FTIR spectroscopy(六色靈芝的FTIR 快速無損鑒別)[J]. Spectroscopy and Spectral Analysis(光譜學(xué)與光譜分析),2002,22(2):226-228.

[11]LIU G (劉剛),LIU JH (劉劍虹),SONG DS (宋鼎珊), et al. Fourier transform infrared spectroscopic study of different parts of wild growing edible mushrooms (野生食用蕈菌不同部位的紅外光譜研究) [J]. Spectroscopy and Spectral Analysis(光譜學(xué)與光譜分析),2005,25(7):1053-1056.

[12]LIU G (劉剛),LIU JH (劉劍虹),YANG AM(楊愛明),et al.Identification of edible mushrooms by Fourier transform infrared spectroscopy(食用菌的傅里葉變換紅外光譜鑒別)[J]. Spectroscopy and Spectral Analysis(光譜學(xué)與光譜分析),2004,24(8):941-945.

[13]SHI YM(時有明),LIU G(劉剛),LIU JH(劉劍虹),et al.Identification of Auricularia auricula from different regions by Fourier transform infrared spectroscopy (不同產(chǎn)地黑木耳的傅里葉變換紅外光譜鑒別)[J].Acta Optica Sinica(光學(xué)學(xué)報),2007,27(1):129-132.

[14]ZHOU ZJ(周在進(jìn)), LIU G(劉剛), REN XP (任先培). Investigation of Boletus speciosum from different areas by Fourier transform infrared spectrometry combined with cluster analysis (不同產(chǎn)地小美牛肝菌的紅外光譜聚類分析研究)[J].Laser&Infrared (激光與紅外),2009,39(11):1158-1162.

[15]LI ZL(李占龍),ZHOU M(周密),ZUO J( 左劍), et al. Infrared and Raman spectral analysis for the corn seed (紅外光譜和拉曼光譜法分析玉米種子的成分)[J].Chinese Journal of Analytical Chemistry (分析化學(xué)研究簡報),2007,35(11):1636-1638.

[16]YAN CQ(晏翠瓊),SHI YM(時有明),LI DY(李棟玉),et al.FTIR characteristics of corn endosperm and embryo (玉米胚乳胚芽的傅里葉變換紅外光譜)[J].Chinese Journal of Spectroscopy Laboratory (光譜實驗室), 2012, 29(6):3525-3528.

[17]SHI YM (時有明),YAN CQ (晏翠瓊),SUN YL(孫艷琳),et al.Study on quality of five hybrid corn kernels by FTIR spectroscopy (五種雜交玉米籽粒品質(zhì)的FTIR 光譜研究)[J]. Spectroscopy and Spectral Analysis(光譜學(xué)與光譜分析),2013,33(11):2945-2948.FENG XJ (馮先桔), et al. Effects of bagging on fruit coloration and quality of nectarine in greenhouse (套袋對大棚油桃果實著色與品質(zhì)影響)[J].Journal of Zhejiang Agricultural Sciences(浙江農(nóng)業(yè)科學(xué)),2011,3:501-503.

[21]YE MZ (葉明智).Preliminary report on bagging of Yingzui peach (鷹嘴蜜桃套袋試驗初報)[J].South China Fruits(中國南方果樹),2007,36(3):78.

[22]SHEN YY(沈玉英),LI B(李斌),JIA HJ(賈惠娟). Effect of bagging on fruit quality of Hujingmilu peach cultivar(不同紙質(zhì)果袋對湖景蜜露桃果實品質(zhì)的影響)[J].Journal of Fruit Science(果樹學(xué)報),2006,23(2):182-185.

[23]HONG L (洪莉), JIANG JY (江景勇),PAN XP(潘仙鵬).Effect of bagging on fruit quality of early-mature nectarine varieties (套袋對早熟油桃品種果實品質(zhì)的影響)[J]. Journal of Zhejiang Agricultural Sciences (浙江農(nóng)業(yè)科學(xué)),2008,2:142-143.

[24]CHEN JJ (陳建軍).Preliminary results of several peach varieties bagging during different growth stages (幾個桃品種不同時期套袋試驗結(jié)果簡報)[J].Gansu Agricultural Science and Technology (甘肅農(nóng)業(yè)科技), 2004, 6: 26-27.

[25]AN XM(安小梅).Effect of different time for putting on or taking off bags on the quality of peach (不同套袋與取袋時間對桃品質(zhì)的影響)[J]. Gansu Agricultural Science and Technology (甘肅農(nóng)業(yè)科技),2003,2:32-33.

[26]KUANG YJ(匡銀近).Effects of girdling and reflective film on fruit quality of“Chunlei”peach(環(huán)剝和鋪反光膜對春蕾桃結(jié)果的影響)[J]. Southwest Horticulture(西南園藝),2000,1:3.

[27]CHANG MH(常美花), ZHANG XH(張小紅), SHI ZJ (師占君). Effects of reflective film on photosynthesis rate and quality of peach and apricot under greenhouse(反光膜對溫室桃杏光合速率及品質(zhì)的影響)[J]. Jiangsu Agricultural Sciences (江蘇農(nóng)業(yè)科學(xué)), 2006,34(4):79-81.

[28]LUO SX(羅水鑫).Preliminary report on application of mulching film and reflective film in early-mature peaches(地膜與反光膜在早熟桃上的應(yīng)用試驗初報)[J]. Fujian Fruits (福建果樹), 2006, 2:27-28.

[29]MU SZ (眭順照), LUO JH (羅江會),LIAO CX (廖聰學(xué)),et al.Improvement of nectarine fruit bagging technique.油桃套袋技術(shù)改進(jìn)試驗 [J]. Journal of Fruit Science (果樹學(xué)報),2005,24(4):396-398.

[30]DOU LJ (鈄凌娟),QIAN DN (錢東南),ZHOU Q (周秦). Application of reflective film in Jinhua Dabaitao peach (反光膜在金華大白桃上的應(yīng)用試驗報告)[J]. Beijing Agriculture (北京農(nóng)業(yè)),2012,33:40.

[31]YANG Y(楊穎),JI RF(紀(jì)仁芬),GU ZX(顧志新),et al.The effect of the reflecting film application on fruit quality of peach (反光膜應(yīng)用對桃果實品質(zhì)的影響)[J]. Northern Horticulture (北方園藝),2012,5:44-45.

[32]ZHANG BB(張斌斌),CAI ZX(蔡志翔),DING H(丁輝),et al.Effect of laying reflective film before harvesting on fruit quality of late-mature peaches(采前鋪設(shè)反光膜對晚熟桃果實品質(zhì)的影響)[J].Jiangsu Agricultural Sciences (江蘇農(nóng)業(yè)科學(xué)),2013,41(9):132-134.

[33]CHEN HJ(陳海江),DUAN HX(段紅喜),XU JZ (徐繼忠), et al. Experiment on improving the peach fruit quality in greenhouse (提高設(shè)施桃果實品質(zhì)試驗)[J].Shanxi Fruits (山西果樹),2003,1:4-5.

[34]TANG XR (唐旭日). Effects of greenhouse environmental factors on fruit quality of nectarine(大棚環(huán)境對油桃果實品質(zhì)的影響)[J].Jiangsu Agricultural Sciences(江蘇農(nóng)業(yè)科學(xué)),2009,37(3):195-197.

[35]CHEN XD(陳修德),GAO DS(高東升),MI QH(米慶華),et al.Effects of different plastic film on fruit quality of peach in greenhouse(不同棚膜對設(shè)施桃果實品質(zhì)的影響)[J]. Chinese Agricultural Science Bulletin(中國農(nóng)學(xué)通報),2009,25(13):254-259.

[36]LI ZY (李中勇), GAO DS (高東升),QIAN J(錢稷),et al.Effects of different light environments on the fruit quality of peach in greenhouse(不同光環(huán)境對設(shè)施桃果實品質(zhì)的影響)[J].Journal of Anhui Agricultural Sciences (安徽農(nóng)業(yè)科學(xué)),2009,37(21):9933-9934,9963.

[37]WANG ZQ(王志強), NIU L(牛良), LIU SE (劉淑娥). Vegetative growth and fruit development of nectarine under protected cultivation conditions(設(shè)施栽培對油桃營養(yǎng)生長及果實生長發(fā)育的影響)[J].Journal of Fruit Science(果樹學(xué)報),2002,19(2):98-103.

[38]ZHAO ZL(趙志磊), LI BG(李保國), QI GH (齊國輝),et al.Advances of study on the bagging’s effecting on Fuji’s fruit quality (套袋對富士蘋果品質(zhì)影響的研究進(jìn)展) [J]. Hebei Journal of Forestry and Orchard Research (河北林果研究),2003,18(1):81-86.

[39]LI XJ (李秀菊),LIU YS (劉用生),SHU HR (束懷瑞). Effects of bagging on color and hormone contents in apple fruits(紅富士蘋果套袋果實色澤與激素含量的變化)[J]. Acta Horticulturae Sinica (園藝學(xué)報),1998,25 (3):209-213.

[40]GAO HJ(高華君),WANG SM(王少敏),LIU JF (劉嘉芬).Preliminary study on mechanisms of putting on and taking of bags in red apples(紅色蘋果套袋與除袋機理研究概要)[J].China Fruits(中國果樹),2000,2:46-48.

[41]WEN S (文帥). Application of techniques on citrus bagging culture (柑橘套袋技術(shù)的研究與應(yīng)用)[J].Academic Perodical of Farm Products Processing(農(nóng)產(chǎn)品加工),2009,7:62-63,72.