Geographic distribution of the eastern honeybee, Apis cerana (Hymenoptera: Apidae), across ecological zones in China: Morphological and molecular analyses

of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information Report
Category:

Short Stories

Published:

Views: 40 | Pages: 10

Extension: PDF | Download: 0

Share
Description
Geographic distribution of the eastern honeybee, Apis cerana (Hymenoptera: Apidae), across ecological zones in China: Morphological and molecular analyses
Tags
Transcript
  Systematics and Biodiversity   4  (4): 473–482 Issued 27 November 2006doi:10.1017/S1477200006002015 Printed in the United Kingdom  C   The Natural History Museum Ken Tan 1 ,Marina D. Meixner   3 ∗ ,Stefan Fuchs 2 , Xuan Zhang  1 , Shaoyu He 1 ,Irfan Kandemir   3 † ,Walter S. Sheppard  3 &Nikolaus Koeniger  2 1 Eastern Bee Research Instituteof Yunnan AgriculturalUniversity, Kunming, China 2 Institut f¨ur Bienenkunde(Polytechnische Gesellschaft),Fachbereich Biowissenschaftender J.W. Goethe-Universit¨at,Frankfurt am Main,Karl-von-Frisch-Weg 2, 61440Oberursel, Germany 3 Department of Entomology,Washington State University,Pullman WA 99164-6382, USA submitted  January 2005 accepted  June 2005 Geographic distribution of the easternhoneybee,  Apis cerana  (Hymenoptera: Apidae), across ecological zones in China:morphological and molecular analyses  Abstract  Thebiogeographyandintraspecificvariabilityoftheeasterncavity-nestinghoneybee,  Apiscerana ,arenotverywellknown.Westudiedthevariabilityofthisspe-ciesinChinausingmorphometricalmethodstogetherwithrestrictionandsequenceanalysis of two different regions of mitochondrial DNA. Samples of   A. cerana  werecollected from feral or traditionally managed colonies in 19 locations of the Chinesemainland,coveringthemainecologicalregions.Workerbeesfromeachsampleweredissectedandmorphometriccharactersweremeasured.Thedatawereanalysedwithmultivariatestatisticalprocedures.DataweresupplementedbypreviouslypublishedChinese samples from the Oberursel data bank and reference samples of   A. cerana from adjacent countries. A mitochondrial DNA fragment containing a non-coding re-gionwasamplifiedandanalysedwiththerestrictionenzyme Dra I.Thisfragmentwassequencedfortwosamples.Forasubsetofsamples,thesubunit2ofthemitochon-drial NADH gene was amplified and sequenced. Morphometric analysis revealed ahigh degree of variation, strongly associated with ecological zones and correlatedwith geographical and climatic parameters. Two main clusters were apparent, onecomprised the bees from the southern tropical seasonal rain forest region, showingstrongassociationstothebeesofVietnam,ThailandandMyanmar.Thesecondmainclusterincludedthebeesfromthetemperatedeciduousbroad-leavedforestregion,the subtropical evergreen broad-leaved forest zone, the high, cold meadow andsteppe region and the North, and showed increasing similarity to the bees of KoreaandJapan.Inparticular,thebeesfromQingzangplateau,onthefringeoftheGanshuprovince,weresetapartbytheirexceptionalbodysize,darknessandpilosity.Therewas no variation in  Dra I restriction patterns within China. Sequence variation of themitochondrialND2regionwasconsistentwithgeographicpatternsofmorphologicalvariation.BeesoftheSouthYunnanregionweresetapartbycharacteristicallybroadabdominal sterna and wax mirrors, with this locally restricted trait transcending themain transition line at the northern limit of the tropical seasonal rain forest region.This northern limit appears to correspond to the separation line between  A. ceranaindica  and  A. cerana cerana . Key words  Apis cerana , biogeography, China, morphometry, mtDNA Introduction Compared with the western honey bee,  Apis mellifera  L., thebiogeographyandintraspecificvariabilityoftheeasterncavity- ∗ Corresponding author. Email: mmeixner@wsu.edu † Present address: Department of Biology, Zonguldak Karaelmas Univer-sity, Zonguldak 67100, Turkey. nesting honey bee,  Apis cerana  Fabr., is not well known. Therange of   A. cerana  extends from Pakistan to Japan and southto India and the Sunda and Philippine archipelagos. Compar-atively few samples and populations have been studied so far and most studies have been restricted to morphometric meth-ods. Based on multivariate statistical analysis of a suite of morphological characters, Ruttner (1988, 1992) suggested the 473  474  Ken Tan et al. recognition of four different groups,  A. c. cerana  from Afgh-anistan through China , A. c. indica  from India through In-donesia , A. c. japonica, andagroupfromtheeasternHimalayamountains,referredtoas  A. c. skorikovi byEngel(1999).Morerecently, molecular markers have been used to study variationwithin  A. cerana  (Arias  et al. , 1996; Smith & Hagen, 1996,1999;delaR´ua etal. ,2000).Basedonsequenceanalysisofthemitochondrial intergenic region between the tRNA for Leucinand the COII gene, Smith & Hagen (1996, 1999) proposedsubdivision of   A. cerana  into three major lineages, a mainlandAsia group that also includes the bees from Japan, a Sunda-landgroup,andaPhilippinegroup.However,resultsfrombothmorphometric and molecular analyses remain insufficient toyield a consistent explanation of   A. cerana  phylogeography(reviewed in Hepburn  et al. , 2001). It is particularly trouble-some that very little is known about  A. cerana  in China, whichincludes the species’ northern limit of distribution in mainlandAsia.China occupies an extremely large territory, encom-passing various geographic regions and climate zones witha high degree of botanical and zoological diversity. WithinChina there are an estimated 2 million colonies of the east-ern honeybee,  A. cerana  (Yaochun, 1993). Of the eight dis-cernible botanical regions,  A. cerana  is primarily restricted tosix: cold-temperate deciduous needle-leaved forest, temperatedeciduous broad-leaved forest, subtropical evergreen broad-leaved forest, tropical seasonal rain forest, temperate steppeand temperate desert. Few  A. cerana  colonies occur in thesmall area of high, cold meadow steppe and none in the high,cold semi-desert and desert regions (Peng  et al ., 1989).Within this variety of environments,  A. cerana  exhibitsextensive variation. Earlier studies separated  A. cerana  inChina into five different subspecies, corresponding to theregions of Hailan, Eastern Yunnan, Southern Yunnan, Aba andXizhang (Tibet) (Yang & Xue, 1982). Later, seven biotypeswere described, which included the Palm forest and mountainbiotype of Hailan, and the biotypes of Guangdong-Guangxi,Hunan, Yunnan Plateau, Northern and Changbeishan (Yang& Xue, 1986). Peng  et al . (1989) reviewed the taxonomicstudies of China and concluded that the patterns of variationfrom the north-eastern parts to the southern parts of Chinacould not be reliably reconstructed from existing studiesdue to methodological and character ambiguities. Thus, thevariation of   A. cerana  in China and relationships to the knownsubspecies of the East Asian mainland remain largely obscure,except for a recent study of bees from Yunnan and Beijing(Tan Ken  et al ., 2003).We investigated the variation of   A. cerana  in mainlandChina through morphometric and mtDNA analyses. A suiteof 38 morphological characters was combined with analysesof two different regions of mitochondrial DNA: the regionbetween the COI and COII gene that includes an intergenicnon-coding sequence, and subunit 2 of the NADH gene. Thegoal of the study was to recognize geographic and geneticvariability of   A. cerana  within China, analyse the relationshipof morphometric characters to ecological parameters and toexploretherelationshipofthehoneybeesofChinato  A.cerana samples from adjacent regions. Materials and methods Collection of bee samples Honey bee samples were collected between 1992 and 2003from nine different regions throughout China (Fig. 1). The19 localities were selected to represent different climatic or vegetation zones, ranging from 50 m to 2700 m in elevation(TableI).Honeybeeswerecollectedfromnaturalnestsorsemi-managed hives, such as log-hives or natural cavities. Movablebeehives have been introduced into some areas recently, butmigratory beekeeping is uncommon.A total of 26 samples were collected. In each locality,betweenoneandsixcoloniesweresampled.Eachsamplecon-tained 30 worker bees, which were preserved in 75% ethanol.Samples were split and one half was deposited in the EasternBee Research Institute of the Yunnan Agricultural Universityof China, Kunming, and the other half in the bee collection of the Institut f ¨ur Bienenkunde, Oberursel, Germany. Preparations and measurements All26colonysampleswereanalysedattheInstitutf ¨urBienen-kunde. From each sample, 15 worker bees were dissected for morphometric analysis and measured according to the meth-ods described by Ruttner   et al . (1978) and Ruttner (1988).Of the 41 morphometric characters listed in Ruttner   et al .(1978), 38 were measured, excluding length of proboscis(No. 5) and cubital veins of the left wing (No. 29 and 30),resulting in 16 size characters, 11 wing angles, 7 colour characters, 3 hair characters and number of hamuli. Measure-ments and colour scaling were performed using a stereomic-roscope and a computer-aided measuring system based ona video system and measuring program ((Bee2,  c   Meixner,2004). Statistical analysis of the data Colony sample means, standard deviation and standard error were computed for each character from the samples by themorphometric measuring program, thus representing estim-ates for the colony. Colony means for China were combinedwith data of 30 previously analysed samples from Yunnan andBeijing (Tan Ken  et al. , 2003). To explore relations to beesoutside of China, data of samples from Myanmar (2), Japan(8), Korea (5), Nepal (4), Thailand (8), Vietnam (17), Malay-sia (2) and India (5) were included. Locations for all samplesare given in Table I. Reference samples were collected frommanaged, semi-natural or natural nests. These data were takenfrom the morphometric data bank of the Institut f ¨ur Bien-enkunde, Oberursel, Germany. Four different methods wereused to analyse the data, each focusing on specific aspects. Todisplay the general morphological relations between samples,data were submitted to factor analysis and sample scores wereplotted on principal component (PC) co-ordinates for visualiz-ation.Subsequently,thestructuringofmorphometricsimilarit-ies was investigated by hierarchical cluster analysis, includingsampling location means for the China mainland samples andgroup means for the bee populations from adjacent countries.Further, a k-means clustering procedure was performed with  Geographic variation of   Apis cerana  in China  475Figure 1  Sample locations in China used in the analysis. 1 – 19 sample locations from this study. 20 – 34 sample locations from Tan Ken  et al .(2003). Location names and geographical coordinates are given in Table I. Circles: tropical seasonal rain forest ecological zone(TSRF); squares: subtropical evergreen broad-leaved forest ecological zone (SEBLF); Rhombus: temperate deciduous broad-leavedforest zone (TDBLF); hexagon: high, cold meadow and steppe zone (HCMS). Membership of each sample in K-means subcluster 1 to 8is marked by x,  + ,  − , I,  \ , /, #,  ◦ within the symbols. For key to Chinese location numbers, see Table 1a, b. Sample locations inadjacent countries within map range are indicated by  + . increasing numbers of clusters until maximal coherence of geographic and ecological zones was obtained. To test thequality of the grouping obtained by factor and cluster analysis,samples were reallocated to their respective groups by dis-criminant analysis. Relations of morphometric traits to envir-onmental variables including altitude, latitude and longitude,were investigated by correlation and regression analysis. Cal-culations were performed using the SPSS for Windows 10.00and Systat 9.00 statistical packages. Molecular analyses Subsets of each sample were transferred to 90% ethanol andshipped to Washington State University where the molecular analyses were performed. Totalnucleic acids of oneindividualworker per sample were extracted according to methods of Arias & Sheppard (1996). Dra I restriction analysis A mitochondrial fragment containing the intergenic regionbetween the tRNAleu gene and the second subunit of the cyto-chrome oxidase gene was amplified using the primer pair E2-H2 (Garnery  et al ., 1993): E2: 5’-GGC AGA ATA AGT GCATTG-3’, H2: 5’-CAA TAT CAT TGA TGA CC-3’. The 50 µ Lreaction mix consisted of 0.8  µ M of each primer, 0.2 mM of PCRNucleotide mix(BoehringerMannheim),1.5mMMgCl 2 (Invitrogen), 1X Reaction Buffer (Invitrogen), 2 U  Taq  Poly-merase (Invitrogen) and 2  µ L of template. The amplificationcycle consisted of an initial denaturation step of 5 minutes at94 ◦ C, followed by 35 cycles of 30s at 92 ◦ C, 90s at 47 ◦ Cand 2 min at 63 ◦ C, followed by a final extension step of 10 minutes at 63 ◦ C.Ten µ L of PCR products were run on a 1.5% agarose gel,stained with ethidium bromide and photographed under UVillumination. 20  µ L of each positive reaction were digestedwith the restriction enzyme  Dra I at 37 ◦ C overnight. Restric-tion fragments were separated on 10% polyacrylamide gels,stained with ethidium bromide and photographed under UVillumination. This analysis was performed with all samplesof the current collection, three additional samples from theprevious collection in Yunnan (Tan Ken  et al ., 2003) andsamples from adjacent populations of   A. cerana  from Thai-land (20), Nepal (2), Vietnam (2), Beijing (2), and Korea (1).Subsequently, the amplification products of two samples fromChina were directly sequenced from both directions using acycle sequencing protocol (Craxton, 1991) and an ABI 310automated sequencer. Sequence analysis of the NADH subunit 2 For a subset of samples (Ganshu1, Ganshu2, Hubei, Sayinpan,Lijian, Hekou, Guangdong, Sichuan), together with referencesamples of   A. cerana  from Nepal, Vietnam, Thailand and SriLanka,partofthemitochondrialNADH2genewassequenced.We used the polymerase chain reaction to amplify a fragmentof about 700 bp. The primers used were developed from pub-lished sequences of   Apis cerana  (Arias  et al ., 1996; Koulianos& Crozier, 1999):AcerND2F: 5’-TTT ATT CAT AAA TTT TAA AC-3’AcerND2R2: 5’-AAA TCT AAT TAA TAT ATA A-3’  476  Ken Tan et al. Location (voucher number) NEcologicalregion Altitude(m abovesea level)Rainfall(mm/year)Mean annualtemperature( ◦ C) Latitude LongitudeK-meanscluster  (a) 1 Ganshu 1 (3255–3226) 2 SEBLF 1200 300 8.00 33 ◦ 13N 105 ◦ 52E 5/62 Hubei 1 (3219) 1 SEBLF 500 1100 12.00 31 ◦ 32N 109 ◦ 23E 53 Hubei 2 (3220–3221) 2 SEBLF 790 1100 12.00 31 ◦ 32N 109 ◦ 43E 54 Hubei 3 (3222) 1 SEBLF 1100 1050 11.50 31 ◦ 41N 109 ◦ 05E 55 Hubei 4 (3223) 1 SEBLF 1650 1020 10.50 31 ◦ 55N 109 ◦ 26E 56 Hubei 5 (3224) 1 SEBLF 1800 1020 10.80 31 ◦ 43N 109 ◦ 08E 57 Hunan 1 (3215) 1 SEBLF 420 1200 13.50 26 ◦ 80N 110 ◦ 85E 58 Hunan 2 (3216) 1 SEBLF 480 1220 13.50 25 ◦ 23N 111 ◦ 56E 59 Hunan 3 (3217) 1 SEBLF 260 1300 14.60 28 ◦ 53N 113 ◦ 56E 510 Hunan 4 (3218) 1 SEBLF 280 1350 14.50 26 ◦ 52N 113 ◦ 52E 511 Jilin 1 (3233) 1 TDBLF 1000 800 6.20 42 ◦ 52N 128 ◦ 31E 712 Jilin 2 (3234–3235) 2 TDBLF 1100 800 6.20 42 ◦ 52N 128 ◦ 13E 713 Ganshu 2 (3227–3228) 2 HCMS 2700 160 2.00 34 ◦ 33N 104 ◦ 23E 614 Guangdong 1 (3230) 1 TSRF 100 1700 21.50 23 ◦ 19N 112 ◦ 48E 115 Guangdong 2 (3231–3232) 2 TSRF 50 1700 21.80 23 ◦ 00N 112 ◦ 45E 116 Guangxi (3212) 1 TSRF 200 1600 21.70 23 ◦ 55N 109 ◦ 22E 117 Huangshan (3210–3211) 2 TSRF 300 1500 15.60 30 ◦ 10N 118 ◦ 21E 118 Hujian (3229) 1 TSRF 400 1500 18.50 26 ◦ 15N 119 ◦ 20E 119 Sichuan (3213–3214) 2 TSRF 450 1300 16.00 31 ◦ 32N 105 ◦ 22E 1 (b) 20 Binchuan (2996–2997) 2 SEBLF 1690 587 18.80 25 ◦ 51N 100 ◦ 33E 421 Huize (3016) 1 SEBLF 1500 822 12.70 25 ◦ 55N 103 ◦ 26E 322 Lijian (2998–2999) 2 SEBLF 2680 938 12.60 26 ◦ 52N 100 ◦ 13E 423 Sayinpan (3008–3009) 2 SEBLF 2250 968 14.00 26 ◦ 00N 102 ◦ 31E 324 Yuangbi (2994–2995) 2 SEBLF 1810 1077 16.10 25 ◦ 39N 99 ◦ 56E 425 Bejing (1370–1373) 4 TDBLF 250 644 11.40 39 ◦ 59N 116 ◦ 11E 626 Caoba (3010–3011) 2 TSRF 1250 827 18.60 23 ◦ 32N 103 ◦ 23E 3/227 Hekou (3006–3007) 2 TSRF 100 1777 22.60 22 ◦ 30N 103 ◦ 57E 228 Jianghong (3004–3005) 2 TSRF 600 1207 21.70 21 ◦ 55N 100 ◦ 52E 229 Kaiyuan (3014–3015) 2 TSRF 1600 813 19.70 23 ◦ 43N 103 ◦ 08E 230 Lushui (3000–3001) 2 TSRF 1950 1159 15.00 25 ◦ 57N 98 ◦ 49E 431 Pingbian (2992–2993) 2 TSRF 1400 1648 16.40 22 ◦ 50N 103 ◦ 43E 1/432 Simao (3002–3003) 2 TSRF 1100 1547 17.70 22 ◦ 43N 100 ◦ 56E 233 Yuangjian (3017) 1 TSRF 1500 801 19.80 23 ◦ 41N 102 ◦ 05E 234 ZhenKang (3012–3013) 2 TSRF 1000 1624 18.90 24 ◦ 00N 98 ◦ 55E 2 (c) Country N Location  (voucher number)  Latitude Longitude Myanmar 2 Taingyr Shan (1301–1302) 17 ◦ 18N 95 ◦ 57EIndia 1 Calcutta (2943) 22 ◦ 34N 88 ◦ 21EIndia 2 Manipur (1232–1233) 24 ◦ 30N 88 ◦ 21EIndia 1 Darjeeling/Lebong (1300) 26 ◦ 58N 88 ◦ 17EIndia 1 Gubudia (1329) 30 ◦ 21N 80 ◦ 32E Japan 1 Kyushn/Fukuoka (946) 33 ◦ 23N 130 ◦ 34E Japan 1 Izuhara-Chou (966) 34 ◦ 14N 129 ◦ 15E Table I.  List of locations and numbers of Chinese  A. cerana  samples per location (N), ecological regions, altitude, yearly rainfall, mean annualtemperature and geographical positions of sample data used in this study. For abbreviations of ecological regions, see Fig. 1 caption.Cluster membership of K-means cluster analysis is indicated. Split numbers show different clustering within locations.  (a)  Samplesfrom China collected and analysed for this study.  (b)  Sample data from China (Yunnan and Beijing) included from Tan Ken  et al .(2003).  (c)  Sample data for reference samples from the Oberursel bee collection.  Geographic variation of   Apis cerana  in China  477 Country N Location  (voucher number)  Latitude Longitude  Japan 3 Shimoagata/Gun (967–969) 34 ◦ 14N 129 ◦ 15E Japan 1 Yamanash (945) 35 ◦ 33N 138 ◦ 40E Japan 2 Tokyo (947–948) 35 ◦ 41N 138 ◦ 40EKorea 5 Seoul/Kyungki (1330–1334) 37 ◦ 34N 127 ◦ 15EMalaysia 2 Malaysia (3236–3237) 4 ◦ 19N 101 ◦ 81ENepal 1 Nagarkot (1076) 27 ◦ 41N 85 ◦ 43ENepal 1 Kathmandu (2526) 27 ◦ 45N 85 ◦ 16ENepal 2 Jumla (1902–1903) 28 ◦ 24N 83 ◦ 33EThailand 1 Chantaburi (1803) 12 ◦ 36N 102 ◦ 07EThailand 2 Bangkok (1322, 1563) 13 ◦ 39N 100 ◦ 31EThailand 3 Khon Kaen (963–965) 16 ◦ 26N 102 ◦ 49EThailand 1 Lampuu Phitsanuloke (1562) 16 ◦ 47N 100 ◦ 16EThailand 1 Fang (1321) 20 ◦ 05N 99 ◦ 30EVietnam 2 Can Tho (2241–2242) 10 ◦ 02N 105 ◦ 47EVietnam 1 Long Khanh (2240) 10 ◦ 38N 106 ◦ 21EVietnam 5 Cuc Phuong (2209–2213) 20 ◦ 15N 105 ◦ 58EVietnam 5 Hoa Binh (1335–1338) 20 ◦ 49N 105 ◦ 19EVietnam 5 Moc Chau (2214–2218) 21 ◦ 20N 104 ◦ 37E Table I  Continued. Reactionswereperformedinatotalvolumeof50 µ Lwitha final concentration of 1X reaction buffer, 1.5 mM MgCl 2 ,0.2 mM of each dNTP, 0.8  µ M of each primer and 2 unitsof taq polymerase. The amplification cycle consisted of 94 ◦ C(60s), 44 ◦ C (80s), 68 ◦ C (120s) and was repeated 35 times,followed by a final extension step of 5 min at 72 ◦ C. Productswere electrophoresed on a 1.5% agarose gel, stained with eth-idium bromide and photographed under UV light. Amplific-ation products were directly sequenced from both directionsusingthecyclesequencingprotocolandanABI310automatedsequencer. Sequences were deposited in GenBank under Ac-cession numbers: AY849558 – AY849569. Sequence alignment and phylogenetic analysis The sequences were aligned using ClustalX (Thompson  et al .,1997) and adjusted manually where necessary. Phylogeneticanalyses were performed using MEGA version 2.1 (Kumar  et al. , 2001). Confidence probabilities for the minimum evol-ution tree (Rzhetsky & Nei, 1993) were calculated using theinterior branch test algorithm of Rzhetsky & Nei (1992). Results Morphometric analysis Factor analysis (FA) of the 38 morphometric characters per-formed on the 107 sample means yielded three factors withhigh eigenvalues ( > 3 . 5) that accounted for 64.4% of the totalvariation in the data. The first factor accounted for 37.6% of the total variation and was positively associated ( r > 0 . 6) withthe size measures length of abdominal terga 3 and 4, lengthof sternum 3 and 6, length of wax mirror, length and width of forewing, length of femur and tibia, length and width of meta-basitarsus, length of cubital vein 1, and with the hairlength.The second factor accounted for 16.9% of the total variationin the data, and was positively associated with the distance of the wax mirrors, the width of the dark stripe of the tomentum,the length of the cubital vein 2, the pigment on abdominalterga 3, 4 and scutellum and the wing venation angle A4. Itwas negatively associated with the width of the tomentum andwing venation angle B4. The third factor accounted for 9.8%of the total variation in the data and was positively associatedwith the width of abdominal sternum 6, the width of the waxmirrorandwingvenationangleO26,butnegativelyassociatedwith wing venation angles N23 and J16.Theplotsofthesamplescoresonthethreeprincipalcom-ponent axes are presented in Fig. 2 (a and b). The sample datafrom China show considerable variability in these plots andsamples from the four ecological zones clearly occupy differ-ent plot regions. Bees from the tropical forest zone (TSRF)were positioned towards smaller sizes (low factor 1 values)and lighter colour (high factor 2 values) than samples fromthe subtropical evergreen forest zone (SEBLF), with onlylittle overlap. Within the SEBLF group some substructurecan be observed, with larger bees mostly srcinating from theeastern regions of Hunan, Hubei and Ganshu, while smaller bees were found in north Yunnan (e.g. Lijian, Yuangbi andSayinpan). Samples from the temperate deciduous broad-leaved forest zone (TDBLF, Beijing and Jilin) were in generaldarker, but similar in size to those of the SEBLF. Two samplesfromthefringeofhigh,coldmeadowandsteppezone(HCMS)in the Ganshu province were set apart, due to their large size,dark coloration of the terga and high pilosity scores, whichexceeded that of any other sample in this analysis.Samples from the TSRF zone showed overlap with thebees of the adjacent southern countries (Vietnam, Thailand,Malaysia, Myanmar, North India), but were lighter colouredand larger, reaching the size of samples from Korea or Japan.Samples from the SEBLF group even exceeded in size thoseof Korea and Japan, but were still lighter coloured.
Recommended
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x