Correlation，Principal Component and Grey Relation Analysis of Sweetpotato Root Biological Traits

Baoqing WANG，Zhaohai DU，Haiyan ZHANG，Beitao XlE，Qingmei WANG，Liming ZHANG.Crop Science Research Institute，Shandong Academy of Agricultural Sciences，Jinan 5000，China；.Agronomy and Plant Protection College of Qingdao Agricultural University，Qingdao 6606，China；3.Shandong Academy of Agricultural Sciences，Jinan 5000，China

Correlation，Principal Component and Grey Relation Analysis of Sweetpotato Root Biological Traits

Baoqing WANG1，Zhaohai DU2，Haiyan ZHANG1，Beitao XlE1，Qingmei WANG1，Liming ZHANG3*
1.Crop Science Research Institute，Shandong Academy of Agricultural Sciences，Jinan 250100，China；
2.Agronomy and Plant Protection College of Qingdao Agricultural University，Qingdao 266106，China；
3.Shandong Academy of Agricultural Sciences，Jinan 250100，China

［Objective］This study was conducted to explore the internal relationship among root biological traits of sweetpotato，as well as the regularity in their formation and differentiation. ［Method］The root traits of 10 sweetpotato cultivars were measured through hydroponic culture in a greenhouse and field survey，and then their correlations were analyzed by statistical methods. ［Result］The root morphological traits of sweetpotato at seedling stage such as projected area，surface area，average diameter and volume processed the highest contribution rate （80.56%）10 d after transplanting，and the contribution rate of root average diameter reached 27.79% 20 d after transplanting.Storage root fresh weight per plant shared extremely significant positive correlations with storage root fresh weight of penultimate node and storage root fresh weight of antepenultimate node，and a significant positive correlation with commercial storage root number，and a significant negative correlation with storage root number of penultimate node.Among them，the correlation coefficient of storage root fresh weight per plant with storage root fresh weight of antepenultimate node was the highest（0.659 5）.Fifteen days after transplanting，storage root fresh weight per plant had significant negative correlations with root projected area，surface area and volume.There was a significant positive correlation between root dry weight and storage root fresh weight per plant 25 d after transplanting. Root dry weight，volume，length，average diameter of sweetpotato seedlings had higher relational degrees with storage root fresh weight per plant.Ten and twenty days after transplanting were important time for the growth and differentiation of sweetpotato roots.In addition，node length and planting depth had certain influence on sweetpotato yield，and direct relationship existed between the seedling root biological traits and storage root yield of sweetpotato.［Conclusion］The results provide theoretical support for standard cultivation and new variety breeding of sweetpotato.

Sweetpotato；Roots；Correlation；Principal component analysis；Grey relational analysis

S weetpotato（Ipomoea batatas L.）is not only China’s fourth largest food crop，but also an important source for industrial raw material，feed and energy，and above 80%of the world’s sweetpotato is produced in China［1］.Sweetpotato has high and stable yield，strong resistance to adverse stresses and wide adaptability.As sweetpotato is propagated by cutting，its ability to rapidly root after transplanting is critical for survival and storage root formation. And storage root number and weight per plant almost determine the production of sweetpotato［2］.Therefore，root growth and development at seedling stage occupy a very important position throughout the life cycle of sweetpotato.Previous studies have proven that the difference in sweetpotato yield is a result of different cultivars，materials，environmental and soil factors［3］.In fact，the storage roots of sweetpotato are mainly developed from the adventitious roots on root primordia of stems and callus［4］.By studying the relationship between storageand adventitious roots of two sweetpotato cultivars Beauregard and Georgia Jet，it was found that the anatomical properties of adventitious roots 5-7 d after transplanting had certain correlations with storage root growth at late growth stages of the two cultivars［5］；and 86%of the storage root of Beauregard and 89%of the storage root of Georgia Jet were developed from their adventitious roots formed within the first week after transplanting［6］.Lateral roots as an important part of sweetpotato plants contribute to the efficient use and absorption of water and nutrient［7］.Moreover，the latest research by molecular biology proved that IbMADS1 transcripts specifically distributed around immature meristematic cells within the lateral root primordia of sweetpotato［8］，and lateral roots probably affected the formation of storage roots by changing the structure of adventitious roots［9］.Togari believed that the balance between cambium development and lignification finally determined storage root number and yield of sweetpotato［10］. Some other studies revealed that the balance between the differentiation of regular and irregular vascular cambiums within roots at early growth stage determined storage root shape and size of sweetpotato［11-12］.Therefore，there may be some relationships between seedling and post-harvest root traits of sweetpotato.

Sweetpotato storage roots grow underground，and thus related measurement only can be achieved after digging them out，which damages its root system.Therefore，in the present study hydroculture was adopted to observe the dynamic changes in root traits of sweetpotato seedlings and the internal relationships between seedling and post-harvest root traits were analyzed using statistical methods，with an attempt to understand the laws in sweetpotato root formation and differentiation，and provide theoretical support for standard cultivation and new variety breeding of sweetpotato.

Materials and Methods

Materials

Ten sweetpotato cultivars mainly grown in Huang-Huai Area Xushu18，Xushu 22，Jixu 23，Jishu 21，Zhengshu20，Jishu 22，Yanshu 25，Jishu 18，Jihei No.1 and Aya were selected in this study.

Experimental design

Disease-free sweetpotato seedlings with similar shape and size were selected and cultured with1/2 Hoagland nutrient solution in a greenhouse.A pump was used to supply air to the solution.Sunlight and air flow conditions in the greenhouse were set the same as in field.Three sets of 20 repetitions were prepared for each cultivar.Hydroponic plastic boxes（45.0 cm long×34.3 cm wide×12.5 cm high）were purchased from Qilipu wholesale market in Jinan City.Twenty potato seedlings were planted in each box，and 30 boxes in total were randomly arranged.Root biological traits were measured on the 5th，10th，15th，20thand 25thd of hydroponic culture.

Field trials were performed in the experimental base of Crop Research Institute of Shandong Academy of Agricultural Sciences.The soil here was light loam，containing 1.83%organic matter，110.2 mg/kg alkali-hydrolyzed nitrogen，58.4 mg/kg available phosphorus and96.9mg/kg available potassium.A randomized block design was adopted.Each plot was 27 m2in area，with six ridges，and each ridge was 6 m long，0.75 m wide and 0.3 m high.Sweetpotato seedlings with similar length and shape were selected and transplanted at a density of 51 000 plants/hm2on June 14，2013. Conventional field managements were carried out.Potatoes were harvested on October 24，2013，while root biological traits were also measured.

Measurement items and methods

Seedling root traits such as root length（cm），projected area（cm2），surface area（cm2），average root diameter（mm）and root volume （cm3）were measured using WinRHIZO （Regent Instrument Inc.，Quebec，Canada），which is an image analysis system specifically designed for root measurement.The fresh weight of seedling roots （g）was measured on an electronic balance.Subsequently，the roots were dried in an oven at 105℃for 30 min，and then at 80℃till constant weight for the measurement of their dry weight（g）.

Root traits in field experiment like callus storage root number，callus storage root weight（g），storage root number of bottom node，storage root weight of bottom node（g），storage root number of penultimate node，storage root weight of penultimate node（g），storage root number of antepenultimate node，storage root weight of antepenultimate node （g），commercial storage root（with weight no less than 100 g per plant）number，pencil root number per plant，pencil root fresh weight per plant（g），lateral root fresh weight per plant（g），and storage root fresh weight per plant （g）were counted or measured with an electronic balance in field.Pencil roots are thin，elongated roots about 0.2-1.0 cm wide and 10-30 cm long at maturity.

Statistical analysis

All data were processed using Excel 2007.Correlation analysis，principal component analysis and gray relational analysis were all conducted using DPS7.05.

Principal component analysis（PCA），which is a statistical procedure that uses an orthogonal transformation to convert a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components，which can control all the original variables［13-14］，was conducted on the root biological traits of sweetpotato at seedling stage and harvest stage.

According to the theory of gray relational analysis［15］，we defined the biological properties of the 10 tested sweetpotato cultivars as a grey system，storage root fresh weight per plant as Y （parent sequence），other root traits as Xi=｛［Xi（1），Xi（2），Xi（3）... Xi（7）］｝=｛［root length，projected area，surface area，average diameter，root volume，seedling root fresh weight，seedling root dry weight］｝，or Zi=｛［Zi（1），Zi（2），Zi（3）...Zi（12）｝=｛［callus storage root number，callus storage root weight，storage root number of bottom node，storage root weight of bottom node （g），storage root number of penultimate node，storage root weight of penultimate node （g），storage root number of antepenultimate node，storage root weight of antepenultimate node （g），commercial storage root number，pencil root number per plant，pencil root fresh weight per plant（g），lateral root fresh weight per plant（g）］｝. After the data were normalized，the gray relational coefficient was set as 0.1 to calculate their gray relational grade，according to which，effects of these agronomic traits on sweet potato yield were sequenced.

Results and Analysis

Principal component analysis on root biological properties of sweetpotato seedlings

As shown in Table 1，5 d after planting，the contribution rate of root projected area，surface area，average diameter and volume of principal component 1 was only 57.58%，while the contribution rate of root length of principal component 2 was 32.24%. Namely，the accumulated contribution rate of these two principal components reached 89.92%，and all these traits of the two principal components were morphological traits，indicating that 5 d after transplanting is a period when roots of sweetpotato seedlings grew rapidly.

Ten days after planting，the contribution rate of root projected area，surface area，average diameter and volume of principal component 1 exceeded 70%，up to 80.56%，indicating that the four root properties of principal component 1 were able to basically describe the root characters of sweetpotato seedlings 10 d after transplanting.Meanwhile，the contribution rate of seedling root dry weight of principal component 2 was 14.78%.As the traits of principal component 1 were morphological characters，and that of principal component 2 was a weight trait，the results indicated that root dry matter began to accumulate 10 d after planting.

Fifteen days after transplanting，the contribution rate of root projected area，surface area and volume of principal component 1 was 60.86%，while the contribution rate of root length，fresh weight and dry weight of principal component 2 was 25.85%.As the traits of principal component 1 were morphological characters，and those of principal component 2 weight traits，the results indicated that both the root weight and dry matter continued to increase at this period.

Twenty days after transplanting，the contribution rate of root length，projected area，surface area and volume of principal component 1 reached 67.64%，while the contribution rate of root diameter of principal component 2 was 27.79%，and the contribution rate of root dry weight of principal component 3 was only 3.32%.The accumulated contribution rate of principal component 1 and principal component 2 exceeded 70%.All the traits of principal components 1 and 2 were morphological traits，and that of principal component 3 was a weight trait.The increased contribution rate of root diameter indicated that root system mainly differentiated at this period.

Table 1 Principal component analysis on root biological properties of sweetpotato seedlings

Twenty-five days after transplanting，the contribution rate of root length，projected area，surface area，average diameter and volume of principal component 1 reached 65.27%，and the contribution rate of root dry weight of principal component 2 was 27.79%. Among the traits of principal components 3 and 4，root dry weight had the largest eigenvalue.The traits of princi-pal component 1 was morphological traits，the traits of principal components 2，3 and 4 were weight traits. The results revealed that root differentiation was almost finished and roots began to expand and accumulate dry matters at this stage.

Table 2 Correlation analysis on root biological traits of sweetpotato at harvest period

Correlation analysis， principal component analysis and gray relational analysis on root biological traits of sweetpotato at harvest period

As shown in Table 2，among the post-harvest biological traits，there were significant positive correlations between storage root number of bottom node and storage root weight of bottom node，storage root weight of bottom node and storage root number of antepenultimate node，storage root weight of bottom node and commercial storage root number，callus storage root weight and lateral root fresh weight per plant，commercial storage root number and storage root fresh weight per plant；there were significant negative correlations between storage root number of penultimate node and storage root weight of penultimate node，storage root number of penultimate node and storage root weight of antepenultimate node，storage root weight of bottom node and pencil root number per plant，storage root number of penultimate node and storage root fresh weight per plant.There were extremely significant positive correlations between storage root number of bottom node and storage root number of antepenultimate node，storage root weight of penultimate node and storage root fresh weight per plant，storage root weight of antepenultimate node and storage root fresh weight per plant；there were extremely significant negative correlations between storage root number of penultimate node and commercial storage root number，storage root number of bottom node and pencil root number per plant，storage root number of antepenultimate node and pencil root number per plant.

As shown in Table 3，among the 13 post-harvest root biological traits of sweetpotato，the contribution rate of principal component 1 traits accounted for 50.87%，mainly including storage root weight of bottom node，storage root weight of penultimate node，storage root weight of antepenultimate node，commercial storage root number and storage root fresh weight per plant，etc.，all of which determined storage root yield of sweetpotato.A-mong the principal component 2 traits，pencil root fresh weight per plant and lateral root fresh weight per plant had positive and larger normalized eigenvectors，both of which were not the traits determining storage root yield of sweet potato.The contribution rate of callus storage root weight and callus storage root number of principal component 3 reached 15.20%.The accumulated contributions of these root biological traits of the three principal components were up to 95.58%，and thus could reflect post-harvest root properties of sweet potato.

As shown in Table 4，the gray relational analysis on post-harvest root biological traits and storage root fresh weight per plant of sweetpotato revealed that the gray relational coefficient of storage root weight of antepenultimate node was the largest，up to 0.659 5.According to their gray relational coefficients from large to small，the post-harvest root biological traits were sequenced as storage root weight of antepenultimate node＞storage root weight of penultimate node＞commercial storage root number＞storage root weight of bottom node＞storage root number of antepenultimate node＞storage root number of bottom node＞callus storage root number＞pencil root fresh weight per plant＞lateral root fresh weight per plant＞storage root number of penultimate node.

Correlations between seedling root biological traits and post-harvest storage root fresh weight per plant

As shown in Table 5，the correlations between root biological traits at different seedling stages and postharvest storage root fresh weight perplant varied greatly.In detail，5，10 and 20 d after transplanting，the root biological traits showed no significant correlations with post-harvest storage root fresh weight per plant.Fifteen days after transplanting，the seedling root projected area，surface area and volume exhibited a significant negative correlation with post-harvest storage root fresh weight per plant.Twenty-five days after transplanting，seedling root dry weight had a significant positive correlation with post-harvest storage root fresh weight per plant.

?

Gray relational analysis on seedling root biological traits and post-harvest storage root fresh weight per plant

As shown in Table 6，the gray relational grade between root biological traits at different seedling stages and post-harvest storage root fresh weight per plant also varied greatly.According to the gray relational coefficient from large to small，the root traits were sequenced as root dry weight＞root volume＞root length＞root surface area＞root projected area＞root average diameter＞root fresh weight 5 d after transplanting，as root volume＞root surface area＞root projection area＞root dry weight＞root fresh weight＞root average diameter＞root length 10 d after transplanting，as root length＞root average diameter＞root volume＞root projected area＞root surface area＞root dry weight＞root fresh weight 15 d after transplanting，as root average diameter＞root length＞root volume＞root projected area＞root surface area＞root dry weight＞root fresh weight 20 d after transplanting，as root volume＞root dry weight＞root average diameter＞root fresh weight＞root projected area＞root surface area＞root length 25 d after transplanting.

Table 4 Gray relational analysis on post-harvest root biological traits and storage root fresh weight per plant of sweetpotato

Table 5 Correlations between seedling root biological traits and post-harvest storage root fresh weight per plant

Conclusion and Discussion

The principal component analysis on seedling root biological properties of the 10 sweetpotato cultivars revealed that the root morphological traits of sweetpotato had the highest contribution rate，up to 80.56%10 d after transplanting，and root average diameter had the highest contribution rate，27.79%20 d after transplanting.The correlation analysis，principal component analysis and gray relational analysis on post-harvest root biological traits of the 10 sweetpotato cultivars revealed that storage root fresh weight per plant shared extremely significant positive correlations with storage root fresh weight of antepenultimate node（r=0.96**）and storage root fresh weight of penultimate node （r= 0.88**），and a significant positive correlation with commercial storage root number（r=0.79*），and a significant negative correlation with storage root numberofpenultimate node（r=-0.87*）. Principal component analysis also showed that roots weight traits decided yield formation of sweetpotato. Storage root fresh weight of antepenultimate node had the largest correlation coefficient with storage root fresh weight per plant with （0.659 5）. Correlation analysis and gray relational analysis on seedling root biological traits and storage root fresh weight per plant of sweetpotato suggested that storage root fresh weight per plant had significant negative correlations with root projected area，surface area and volume 15 d after transplanting，a significant positive correlation with seedling root dry weight 25 d after transplanting. Moreover， root dry weight，volume，length，average diameter of sweetpotato seedlings had higher relational degrees with storage root fresh weight per plant.Twenty-five days after transplanting，

Table 6 Gray relational analysis on seedling root biological traits and post-harvest storage root fresh weight per plant

Ten days after transplanting，root morphological traits of sweetpotato had the highest contribution rate，up to 80.56%，indicating that it was a period when root system grew rapidly.Twenty days after transplanting，root average diameter had the highest contribution rate，27.79%，indicating that indicating that the differentiation of roots especially the storage roots were completed.Twenty-five days after transplanting，root dry weight had the highest contribution rate，up to 27.79%，indicating that dry matters began to accumulate in roots.Therefore，10 and 20 d after transplanting might be important time for the growth and differentiation of sweetpotato roots，which agreed well with previous studies［16-17］.At the two periods critical for sweetpotato seedling growth，more attention should be paid to avoiding the adverse influence caused by environmental factors such as drought，low temperature，waterlogging on root growth and differentiation.

At harvest stage storage root fresh weight per plant shared extremely significant positive correlations with storage root fresh weight of penultimate node and storage root fresh weight of antepenultimate node，a significant positive correlation with commercial storage root number，indicating that sweetpotato yield per plant correlated with the length between earth surface and storage root growing nodes.Sweetpotato yield will be increased when nodes close to earth surface root and grow storage roots earlier.However，storage root fresh weight per plant and storage root number of penultimate node shared a significant negative correlation，suggesting that more storage roots on penultimate node would reduce sweetpotato yield.However，the positive correlation between storage root number of antepenultimate node and storage root weight per plant was not significant，indicating that the storage root number on different nodes had different influence on sweetpotato yield，while storage root weight might have greater influence than storage root number on sweetpotato yield，which was also proven by principal component analysis and gray relational analysis on post-harvest root biological traits.The weight traits of storage roots especially on antepenultimate node contributed most to and also correlated most with sweetpotato yield.In fact，the distribution of root biological traits on nodes were rarely reported in previous studies.In actual production，appropriate planting depth will improve sweetpotato yield.Besides，sweetpotato cultivars those grow larger and more storage roots on nodes close to earth surface can be selected for breeding.

By analyzing the relationship between seedling root traits and postharvest storage root yield per plant，we found that root projected area，surface area and volume had significant negative correlations with storage root fresh weight per plant 15 d after transplanting，and seedling root dry weight had a significant positive correlation with storage root fresh weight per plant 25 d after transplanting，indicating that appropriate root growth and differentiation would contribute to the increase of sweetpotato yield.Seedling root dry weight，root volume，root length and root average diameter highly correlat-ed with storage root fresh weight per plant，and all of the four traits were related to horizontal and vertical root enlargement.So far，sweetpotato root system has rarely been studied，and the relationship between seedling root traits and post-harvest root traits has not been reported.

To sum up，much attention should be paid to the management of seedling growth in production to coordinate between planting depth and node length.Sweetpotato cultivars those root and differentiate earlier should be selected for high-yield cultivar breeding.The internal relationship between seedling root traits and final yield of different sweetpotato varieties from different ecological zones will be analyzed by measuring more indices to improve our research.

References

［1］WANG BQ（汪寶卿），XIE BT（解備濤），WANG QM （王慶美），et al.Research advances of endogenous hormones and chemical regulation on sweet potato（Ipomoea batatas L.）（甘薯內(nèi)源激素和化學(xué)調(diào)控研究進(jìn)展）［J］.Shandong Agricultural Sciences （山東農(nóng)業(yè)科學(xué)），2010，1:51-56，62.

［2］WANG Y（汪云），CHEN SY（陳勝勇），LI GK （李觀康），et al.Advances in research on drought resistance in sweetpotato （甘薯抗旱性研究進(jìn)展）［J］. Guangdong Agricultural Sciences（廣東農(nóng)業(yè)科學(xué)），2011，38（11）:35-38.

［3］KAYS SJ.The physiology of yield in the sweet potato ［A］.In:Bouwkamp，J.（ed.）.Sweetpotato products:A natural resource for the tropics［C］.Boca Raton: CRC Press，1985:79-132.

［4］BELEHU T，HAMMES PS，ROBBERTSE PJ.The origin and structure of adventitious roots in sweetpotato（Ipomoea batatas）［J］.Australian Journal of Botany，2004，52:551-558.

［5］WILSON LA，LOWE SB.The anatomy of the root system in West Indian sweet potato［Ipomoea batatas（L.）Lam.］cultivars ［J］.Annals of Botany，1973，37: 633-643.

［6］VILLORDON AQ，LABONTE DR，F(xiàn)IRON N，et al.Characterization of adventitious root development in sweetpotato［J］.Hortscience，2009，44（3）:651-655.

［7］CASIMIRO I，BEECKMAN T，GRAHAM N，et al.Dissecting Arabidopsis lateral root development［J］.Trends in Plant Science，2003，8:165-171.

［8］KU AT，HUANG Y，WANG Y，et al.Ib-MADS1（Ipomoea batatas MADS-box 1 gene）is involved in tuberous root initiation in sweet potato（Ipomoea batatas）［J］.Annals of Botany，2008，102:57-67.

［9］VILLORDON AQ，LABONTE DR，SOLIS J，et al.Characterization of lateral root development at the onset of storage root initiation in ‘Beauregard’sweetpotato adventitious roots ［J］. Hortscience，2012，47（7）:961-968.

［10］TOGARI Y.A study of tuberous root formation in sweet potato［EB/OL］. Bulletin of Natural and Agricultural Experimental Station，Tokyo，1950，68:1-96.

［11］LOWE SB，WILSON LA.Comparative analysis of storage root development in six sweet potato ［Ipomoea batatas（L.）Lam.］cultivars.1.Storage root initiation，storage root growth and partition of assimilate［J］.Annals of Botany，1974，38:307-317.

［12］LOWE SB，WILSON LA.Comparative analysis of storage root development in six sweet potato ［Ipomoea batatas（L.）Lam.］cultivars 2.Interrelationships between storage root shape and yield［J］.Annals of Botany，1974，38:319-326.

［13］TANG QY（唐啟義），F(xiàn)ENG MG（馮明光 ）.DPS data processing system（DPS數(shù)據(jù)處理系統(tǒng)）［M］.Beijing:Science Press（北京:科學(xué)出版社），2007.

［14］XU KX（徐克學(xué)）.Biological mathematics（生物數(shù)學(xué)）［M］.Beijing:Science Press（北京:科學(xué)出版社），1999.

［15］DENG JL（鄧聚龍）.Grey theory and method in agricultural system （農(nóng)業(yè)系統(tǒng)灰色理論與方法）［M］.Jinan:Shandong Science and Technology Press（濟(jì)南:山東科學(xué)技術(shù)出版社），1988.

［16］Jiangsu Academy of Agricultural Sciences，Shandong Agricultural Sciences（江蘇省農(nóng)業(yè)科學(xué)院，山東省農(nóng)業(yè)科學(xué)院）.Sweet potato production in China（中國甘薯栽培學(xué)）［M］.Shanghai:Shanghai Science and Technology Press（上海:上海科學(xué)技術(shù)出版社），1984.

［17］RAVI V，INDIRA P.Crop physiology of sweetpotato［J］.Horticultural Reviews，1999，23:277-339.

Responsible editor:Qingqing YlN

Responsible proofreader:Xiaoyan WU

（Continued from page 478）合力和遺傳力分析）［J］.Acta Agronomica Sinica（作物學(xué)報），2005，31（6）: 784-789.

［23］LU YG（盧永根），ZHANG GQ（張桂權(quán)）. Specific affinity genes and ideas on their utilization in hybrid rice breeding（特異親和基因及其在雜交水稻育種上利用的設(shè)想）［J］.Fujian Science and Technology of Rice and Wheat（福建稻麥科技），1992，10（1）:1-4.

［24］ZHU Z（朱鎮(zhèn)），ZHAO L（趙凌），ZONG SY（宗壽余），et al.Analysis on combining ability of main characters in two-line japonica hybrid rice（兩系雜交粳稻主要性狀的配合力分析）［J］.Acta Agriculturae Jiangxi（江西農(nóng)業(yè)學(xué)報），2006，18（1）:11-14.

Responsible editor:Xiaohui FAN

Responsible proofreader:Xiaoyan WU

Cambridge Scientific Abstracts（CSA）

In 2011，Agricultural Science&Technology’s full-texts have been included by Cambridge Scientific Abstracts（CSA）.CSA is a retrieval system published by Cambridge Information Group.With a history of more than 30 years，the company is an important publisher of publishing abstracts and indexes with agent centers in Britain，F(xiàn)rance，Australia，Japan，Netherlands，New Zealand，and Hong Kong China.The products include traditional printed abstracting and indexing journals，a variety of electronic database which are retrieved through remote on-line and CD-ROM.Recently，the CSA launched network based （IDS）services，which can retrieve hundreds of databases provided by CSA and CSA publishing partners.

For more information on CSA Full Text please sign in http：//www.csa.com/

甘薯根系生物學(xué)性狀的相關(guān)性、主成分和灰色關(guān)聯(lián)度分析

汪寶卿1，杜召海2，張海燕1，解備濤1，王慶美1，張立明3*（1.山東省農(nóng)業(yè)科學(xué)院作物研究所，山東濟(jì)南 250100；2.青島農(nóng)業(yè)大學(xué)農(nóng)學(xué)與植保學(xué)院，山東青島 266109；3.山東省農(nóng)業(yè)科學(xué)院作物研究所，山東濟(jì)南 250100）

［目的］研究甘薯根系生物學(xué)性狀的內(nèi)在關(guān)系，了解其形成和分化規(guī)律。［方法］通過溫室水培和田間調(diào)查相結(jié)合的方式，利用統(tǒng)計學(xué)方法研究了10個甘薯品種的根系性狀。［結(jié)果］栽后10天，甘薯苗期根系的投影面積、表面積、平均直徑和體積等形態(tài)性狀貢獻(xiàn)率最高且達(dá)80.56%；栽后20天，根平均直徑的貢獻(xiàn)率最高且達(dá)27.79%。收獲后，單株塊根總鮮重與倒三節(jié)、倒二節(jié)塊根重量呈極顯著正相關(guān)，與商品薯數(shù)呈顯著正相關(guān)，與倒二節(jié)塊根數(shù)呈顯著負(fù)相關(guān)，與倒三節(jié)塊根重量關(guān)聯(lián)系數(shù)最高且達(dá)0.659 5。在栽后15天，苗期根系的投影面積、表面積和體積與單株塊根總鮮重之間呈顯著負(fù)相關(guān)；而在栽后25天，苗期根系干重與單株塊根總鮮重之間呈顯著正相關(guān)，苗期根干重、根體積、根長和根平均直徑等性狀與單株塊根總鮮重關(guān)聯(lián)度較大。栽后10天和20天可能是甘薯根系生長、分化的重要時間節(jié)點(diǎn)，節(jié)間長度和栽插深度影響產(chǎn)量形成，苗期根系生物學(xué)特性與收獲后塊根產(chǎn)量之間有直接關(guān)系。［結(jié)論］研究結(jié)果為甘薯標(biāo)準(zhǔn)化栽培和新品種培育提供理論支撐。

甘薯；根系；相關(guān)性；主成分分析；灰色關(guān)聯(lián)度

國家甘薯產(chǎn)業(yè)技術(shù)體系（CARS-11-B-11）。

汪寶卿 （1979-），男，山東臨沂人，副研究員，博士，主要從事植物激素生理和化學(xué)調(diào)控方面研究，E-mail：xb970607@163.com。

，研究員，博士，主要從事甘薯生理和生物技術(shù)方面研究，E-mail：zhanglm11@sina. com。

2014-11-25

2015-02-14

Supported by National Sweetpotato Industry Technology System（CARS-11-B-11）.

.E-mail:zhanglm11@sina.com

November 25，2014Accepted:February 14，2015