s 540 .8 .C4 58 IfI5TOilU 5 5,IY Subcontracl Agl'ftment (ur tbe t:x«ution or USAIDISADCISARRNET Granl No. 69O-G--OO---99-00258-00 betwftn Ihe huemltiooallllstiUUe (or Tropicl l Agric:ultu~, UTA. _lid tbe ¡ntem.rional Center far T ropical Agritulture, O AT Final Report U"I[/"I,' k .LH'¡IA.:: on Presemed by: lhe lmemational eente!" for Tropical Agricu lture, CIAT Presente!! 00: lhe l mernational Institule for Tropical Agriculture, TTT A December. 2003 g~IT'~~ ------ o CQNTENTS 1.0. lotroduf lioo . ...... ........ .•. •.•. .. ... .. . ... ......... ..... ....... ..... .. ... .......... . 2.0. Rcporl 00 a("II ~lties underlaken In ¡he liubconlntC I agr~OJent . 3.0. Work plan: Wha l "'as done? \ .... hllt \fIlOS ¡ ' cco01pli~hed? ........ ... . J .1. 1lIemo: 1. Estahl ishmcm uf a pri>·atc:·puhlk conSOfllum 10 Sllppor1 c.:oss~ v:l. research und yslem 57 Table 6 Cassava Icaves trials in Malawi-2ool-2oo2. Detopping harvest system W Table 7 Yields of ca,sava produclion systems in 1"10 growing cycles. Malawi, 2001 -2003.. ..... .. .. .. .. .... .... . .. . ...... . .... 62 Table 8 Sweet polato fenilization lrials in Tanl.ania, 2001 ............ .. .. .. ... .. . Table 9 Sweet polato fertilizulion lria ls in Tanzania, 2oo! Table 10 Cassava Icaves lrials in Tanz3nia, 2002 .. Table 11 Cassava roo! production trial, T¡mzania, 2002. LlST OF FIGURES Figure I Performance uf broi1er:s fed "ilh cas..'ilI\"&. Tanzunia. 2002 ... , ....... .. ... , .. )' igllre 2 Performance uf ml lklng cows in Mal3wl red wim cassava slluge .. Figure 3 Web p~ge of the Informalhm Syslem on Casuva POSlhar\'esl Manage~111 IUlU Pl\lCessing ......... .. , ....... ............... , ....... .. .................. ,. , Figure 4 Cas~ava Icaves prWucl'on Inal in Mal3wi. Ratooning 5)'Slcm, towl )'idu in fi" ... harveslt,..... . ... . . . ,.. ........ . ..... , .. , ..... . tl¡.:urt' 5 Cauava leaves pruductlon !nal in Malaw,. Ratooning ~ysICIO, total yie lu in I¡ve har>'CSIS, ........ .... ..... ..... ... ....... , .... .. , .. , ........... " Flgunc' Cassava leal'CS Prodo.",lion tria! In Malaw;' Detopping s)'slc m. 100al ylc ld in fj vc harvcslS ........ ..... , ...... . , ....... ..... . , ... ............. .... .... . Cassava Icavcs producllon !nal in MaJawi. DetOpplOg syslem. tOlal yicld ," fiyc harvcslS ...... ,., ................ . ,.' , ............. ...... ......... .. Figun- S Ylclds of c~va pmductjoo 5yslcms In lwo grov.lng cycles Malawl. ~ .. " .17 " " 2OO 1-200J ........... " ............. '."." .......... .. ................. . , .. , 62 SWetl pollitO rcrtilizuuoo lrial, TallZImia, 2001. ... Swee¡ potalo fcnil ization tri al, Tanzania. 2002.0 .. flgure 11 Cassava leuI'es pmduction tr101 s m Tum::;nia. 2002 .. Figurt' 12 C.SS¡lVP roots prooucllon lIlal in Tanzaru ll ....... .. '" Subc:ontn.fl ;\g~lUent ror the EXl'Culion vi" USA 1000A.OCISAilItNET (jnull No, 690-G-OO-99-«l2511-OO bd"'~1I Ihe Internat iona l Inslilule rur Tropifal Agricultu !"r. liTA. IIl1d Ihe IlIterna lional C~nttr ror TropiclIl A ¡:r1fultu 1'(:. e l A T I.O.lntroduclion Thc Soulhcnl Afri~~n Rool Resean::h Nel"'urk (SARRNET) $tek~ 10 incrc:<= lOCOmc aOO impn:we Ihe wclfare of!he rurJI poor by Incrc:asing small farmer caS~3V3 and sweet poi::uo prOOuctivity thruugh: ~ The adopUOII or Impmved "arielles ami j)f:ll:lices, :;;. The stn::ngthening of national mol cmps researeh capabil ilies.1lJKl .. Cummerdnhz8tion 10 iocressc lIdded v:llue 10 the crops. The suoconlrJCt agl«mem between lhe Inlemalional lnstitule uf Tropical Agri~ulture (liTA) and Ihe Intemariona l Cenlcr fOT Tropical Agricullllre (CIATl is 10 proVlde pmfesslOIlUl ¡nd O/heT relsted servkes 10 meel lhe overall ObJcclive of SARR NET. describcd aboye. ClAT 1'.'11 1 pro~'lC1e lhese ser\'iccs \hmogh liS uwn Rural Agmcllterpri~ Oevclopmenl ProJCCI 1100 Ihmogh 11.5 membership uf lhe ulill American Consonmm for C~ss.: ... a Rescarch and Oevclopmcol (CLA YUCA), Thesc servlCes wlll conlnb\!lc 10 :.chlc'·mg Ihe ~peci fic obJect l'es of Ihe followmg rOUf lheme arcas: Theme 1: F.~lahli.~h of s priv3re-publil: .:unsortiurn 10 support rescareh and deve lopmenl uf swccl pollllO and c as~a~a in ut le.1~t Olle coumry. Themc ti : De\uiled unde]"'!;l.1nding of the Subscclur ~od markeling opJXlrtllo ities 10 increas.e awan:ness amoog Slakeholders of polenlipl opportunil leS for lhe crups I!I lhe region. Thcme 111: Sustalnable. apprupriate pn.K.~sl!lg technoJogies adoplcd by farme]"'!; and Indu~tnaJ process.ors. Themc IV: Sl~keholdcrs Ir:tined in ncw skilJs lO ~upport lhe murkel drj\'en strategyal reglOoal level. 2.0. Repurl on acli\'ilies underlllken in Ihe suocoolraclllgrocrncnl FoIIOWI!lt; !he ddinition 01" o ""01'" plan for lmplemcmllllOll of lhe UTNCJAT SUOCOnlr:lC1 agrecmcnl. ClAT and C LA YUCA persoonel parlicipated io various acljv lILe~ r rable 1), Oased on rhe resulls 00311led 111 lhese acl1Vl lies, a descnplJon I~ p"ell 10 eocn of Ihe four lhcmcs mcluded 111 lhe work plan. wllh cxplanalions 3bout lhe :.clh'lIles Implememed. Ihe results ochlevcd and lOme con.::tusIOIlS, • T.ble l. Summary o, nutior Inttn-r:nlions ~r CIA T & C LA YUCA COlUultaMy T~m. Puiod No.-r:mber 2000 ,o MalTh 2003 AnIoto J 1101 • 'MUGa CLA \'UCAlClA T pt.-.I IIo>'Oh"td ,....... Cou..trIa . Iotud oo. " '_1 J."OU'" ,n T ........... 8",,,,,," OspoNl and Tanuno&. MI""-' (I'dItuory 11 lo> JWWl BwnraJll Ind Keny> " U_ ........ , , ....... B«N o.p. ... " Jw¡jn 811,ltlJll " " ....... :Dd as a 'Iechnoloer Atcllda nf lhe wurlo.:$hl'lll "'H) c~arinl house' rOO' cuu.1 and ,,,'ttl poUlO prodl.lC'lion and pn;I<'e>"nS ~,- 2 Conta<" ""IIh M.I ....... " SARRNETT .... m SpeclflC in'crc .. o Idtnolr..,,¡ IndU .... I;IbI .... 'idl Informouo".ruw InoCrw In ""ng co,,,"o/,,,,«, CJATI CLAYUCA AClion pl.", okveloped .""' ... ;eri ........ nd lI«Ib of ¡)(Un,,", p;Yu,c" poIl'O produm • nd local priv.", lOO InfL>rma.tion on .... hnolofr "1" ....... .1Id lICI_>n 111- r ... prl.lpOWd at:UOIIf. pub lic p.ft ....... d.thverod l. u...ta<1> wllh Tan KknlirlCd ....... i"~ CIA T I('"I.A YlJ(" A _ Adl"" pla"" IkvdupcJ dwa.:;ICt,,' ... ·J .nd nocd$ 01 ~cm .. 1 ¡w111efS ,_ in \IIln, ..... va/,....." . 1Id Ioc.l p,.,vat. , lid - Informa'inn "" l .. dl",,~nl.Y opI""'" .1Id ... \I0Il plllIIli fQr p ptlbhc paft""'" dehvered 4 FormuIOl"'" or tltc: C""5O<1lum SARRNET T .... m - Drlft O~ofdr~ft n .... " .v.il .bl. , 1'heme l . elta/Jlish a pri''Qtt-pubfic conloroum 10 IUpparf ~u,,~h "lid dl!.·t/opmt nt of SM·ttf p%/" ond ,aifa .... in rú lttlSl ane (aunt'] 1001-1001 (oontinuatian) Arth1Iy .... hiel P.nldputs E.J;pec1N "".pu. 11NI1nl1o. (.) ro • ...,lIIto -, 1'rqIa"'''OIl Q( 111>1 ..... .)1 0fI1'OCII mlp SARRNET T<>:hnnlocY nl"in .. o odc ... ¡rlCd ..... , .... ,"! L >I or .c;:hroIo~_ n~;bbk arrordJ.., 10 pnx~ss",* (~.uloclOU. plCCS. 1a)'OlJU. CIATICLAYUCA 10 speciroc uq~ ~nd I ... er ....... ca.:h Ppmenl Oo>oloay pacb~ avall.-bk fur "'ICCS. ~~Ioc.. ele Cal$a"O fIour fu< o ... mal roeto!'lIg. lIo",h ard c'oquettc. fu< human ""'''''''opI_. ard .. la .. ti Meollnl ro< Iluro: h,"g lhc SARRNETT .. m Al ....... "". ........ lI1'um C>I~bh.-.hcd • ..J Offo.:ial doi:umcm> h~e A<:.>. by l.", .. of.he CO<>SQrt,urr>. 01 needed . in lile." """ C)ATICLAYOCA ope ... "nl ,~""""'¡"um. );s."f acuvILio-sI WO<~ plan. c"""'')' llId local >"""" llId ""blic panncTh 7 1"",lemcn .. '.m. Mn"i'ntin¡¡.oo ~¡\RRNIoI Tcam FeaJb.d Infu'mal'''" "" ¡>oOj«1 .. es"lIs PrOIl«'" rcpnn.< .• nnlllll "porl'. SARRNET Eva l","on CIAT/CLAYUCA "".,I.hle W"bPale •• ROOT!S . <",",,;al documenl' (o..-Cl1l n8. COnllnUOUJ pmcc.5s1 Feedtack ,"Ior",,"on deh"",aJ~, prepaocd . .... ~eh.ojd." M~~" inf",n\lti"" avaUable Ih rou¡h SARRNET W"b p.¡¡e. IWOTS. 10 ... ld."ldtf~ includ¡lIg SC .nd p'"'''''''' , 3.1.3 Resullli ilc hleved : a. Cassam UIIJ .\ ..... eel poIa/Q .... Qrbh"p in Mala .... ' TIle camlVwsweel pol&10 workstlop was reaJized in Muy 21-23. 2001. SA RRNET ¡.ineClu Variou~ document5 Wilh InfOrmalion UboUI <.Iiffcrem lcchnologie~ nnd quotalions fOT processlnS cquipmenl w~~ ~sscmbled nnd dehvered la SARRNET IlCrsonnel in Tanzania and Malawl. These informatlOn package) "'cre deJivered in lhe fom of wflnen documeols, catalogues. cleclrunic files (Word documcms and power puil\! pre.l.entations) . and comp""' l d' sks ¡... A documenl <.lo lhe use of cass~v~ in fIOu ltry nlllrilion, onginall), in Spllllish ..... ·as lranslated into English ~nd deh\"crcd lo SA RRNET as a pl,nted document and also in electronic form (PDr fonn~l) ;... A documenl un lhe use of cassava in (\a,ry fccdmg. ongina!!)' in 1'hlll, was translated into English and dehvcrcd to SA RRNET as a printcd document and airo in elec[ronic fom (PDF funn~l) ;. The first draft of [he translation to Engli~ h of Ihe book "Uso o.Je la Yuca en lo Alimentacion Ammar·. origmally publishell in Spanish. W3S com:luded In OclObcr 20m. The draft docurnent ,,"¡sS oJclivcrcd tO FOODI\'ET. wh,ch has agreell 10 finance lhe ne",1 s[ep of thls work.:, scienl ilic editing. f .\Ien'ng lor Iu""ching lile Consonil/m. jI nude-d. in <1' 1e-N (blscuil. baker)' aod nour), 4 brewers, 2 0011-1000 industries (chemical s and text iles) and I food mspecl10n induslry. 11lc complete lisl 1$ shown in Table 2. CII ~sa\'1I n our rn r d lrt'('1 ~1I 1t. tIl Clln~U !!Iers. Analysis of Ihe 10ronn~ lI on obtai ned sllowcd Ihal 10 lhe shon tI:= lhe markel ",i1h lhe grcatest potential for growth is high qUidil)' cassava Oour ror direct sule lo CQl\SI.uncn. 1111s product i5 already 00 the mllI'kel. and allhough \he lotal "ohm\e of sales is prc:semly lo\\', infonrulOts manifcstcd an incn:asing demand for D qUlllil)' prodOCI. Cassava noor is prefeTTC'd nour fO( sorne elhole gJOUps. Al pR'scol , CllSSllva Oour retai l price is USS 50011. ",hieh compare;<; favourably wilh a price of USS 300f1 for maize Oour. More detailed !!ludies. 5peclfic 10 lhis producl. will need 10 be undcnaken 10 ascer1llJn Ihe likcly 10lII1 "olume Ihlll could be dcmanded lo Ih i5 mari;:ct. Cassa va 0 011 r for Ihe food Imlll sl r )' . lncreas.ng supply of tassa~a nOI,lT fUI' direcl So'de 10 consumers, lhrough supermarkc ts and oIhcr outlets. coold open up lhe opponuoily of bn::.wng into the flJlXi mWll,lr:loC luring industry (biscuil, bread nnd pos~i hly OIhe .. thal use llour bascd prOOl,lcts). l"le res t was shown by food procc5S0rs, bUI pre~ently the", is insufficicol producllon of high qualil}' flour al J compelilivc pricc 10 meel thei r rc:qui",menIS. Whc~1 Oour price5 for industrial food use in Tanzania ~re now US$ 220- 2601t. Cassa~¡¡ noor " 'ould need lO be priced helnw Ih,~ In be ahle lO compele. Sinee Ihe wholculc price of high quali l), nOllr for direcl nk 10 1:0nsu111('1'$ is US$ 40011, and demand is Incn:as;f\g. mere is 00 il1Centi\'e al p .. esent for nour processors 10 seU to food manufoclOI"CT1'. As lo Ihe case of t1lSSaVa n our for d"«l $lile 10 consuTl"lCni. a specific slOd)' will ha,'c 10 be undcrtakcn lo defi ne mon: prcciscly Ihe le"e] u f dcmam.l in Ihe food if\doslry. 1111$ ha! 11,1 be 3CCompanied by ool1aboralion wllh lile loduslnes Ihemsel\'C:s 10 ascen alO opt imum leve ls of Callsava nour in Ihe.r ronnulal lon5. Incre3.'lCs 10 C3.~..a\·a productivi ty and more efficlenl proccssiog wiU be required 10 impro"e compctih vencss uf floor prodUClion. , Se<, Cl.A YUCA. OAT, SAlUtNET. lrf A. CIP. rool .nd WO~ Repon ,,,, lOe SUtocOlllT1lCt ",,","mellt f(O' lile e"""' ......... of USAJOISADCISARRfIo'ET {".rw11 NiI. 69Q.O.oo.99-OOlS8-OO be' ...... n ,he fn,ematÍ 1m t.ot. al ..te! .- IOlherN Ihn .. i/I IoID.hn _,., .. 10 I..lATIt.'t.A YUCA 1!Id ~ .. _:11. crup *""r .... (I..ulatok) nIlU. F- ,n Tal\UlUlO ....:1 t"Iot of I.'OA ""\lCtIllC Ior ud. M ..... ; Prcl;"'-r _ llf'IK'I\Q ro. udo ... - nur\eI ~) dlobonlcd 1m acto COOIIIII)' (pend ..... ",,11 b< ""'" "" "'" .....tb I)j "'" "" ""111 fodd Inall) L Rnww -.:1 .... ,... ,,1.-...-- C-..... "'_IPhIn el al J MatL.:l du.ft r.. .,... •• n. ... and """"""I'ry da ....... d .. b l'" ed ......... _.It)"~1y .""'" ~.ll) .f'.:nd,nl fu< .aou .. ¡>I1ldll. .... ' ''«1 o""'''yl 1''''''''' in Tana/IIoIl!Id Mil .. " 1.I;Lo,*, ,. The contact, estublished in Tanz~niu with I !lCctors interested in cas~a .. a and sweet potato incJuded rour brcwcry compames. Although a cOllplc of them e~pressed imerest in ~onducling Irials with cassava nour, during Ihe perind covcred under this agr~emcnt, no testing w:J.1 conducted wilh them. Table Z. Lisl of eompanies ,·isi tcd in Tanzania. Ca_saya chips amI nour ror Ihe dairy and nouttn fero sector. B01h the poultry and dairy sectors are growing as Ihe demand for chicken and dairy prodUCIS increases. This pmvidcs opportunities for dry cass:¡va and sweel polmo chips to replace partiaJly maile in balanceo feed fonnulations and d¡rect feeding of silage (rool~ and Icaves and vines) or dry producl~ \O dair,' catt le. Present malze usage 10 animal feed manufacture (bascd on lhe induslries visiled) is estimated al 150.000 tlyear. The CUITent price of maize is US$ 16011 (hut Can fa ll as low a~ USDS 90ft), This dmmatic faH in price can be cxplaincd ea,i l)' ir we consider !hal the maize crop is a commodily highly suhjecl to polle)' i'~ue ... . " Sorne years lhere i, plenly o f imponed maize a'·ailable. wheRllS in other ~anJ lhe imponed volume$ are lower and local prices lemJ 10 increase con&idcrably . Assumlng an 1I"erage 20'1> replocemenl of m3.lu in lhe r~liuns. Ihere 15 11 potenl;al dem:lIld for 30.000 I of cass:l\'a chips, PiIOl. lrials undcnaten by lhe proJl:C1 have sho"'n thal US51Iva ch¡~ c~n be produccd al US$ 8011 chips. wh ich is oompetitive allhough wilh a very n.:llTOW proril margino AChons needl:d 10 capture lhis polenlial martC:1 dc:rnand Inc lude lhe disseminallon of chl pping :md dryi ng t~hnolugy and funhcf lrials 10 Oplimlse and correet dericiencits in lhe use uf cllssava and swecl poIato in balanced fttd nuions. For dalry producers. Q ll'IOI.onzed combinoed chopper--chlpper for le:l\'cs has been mtroduced bul needs 10 be: rerined (buih through SARRNET-T anzama and a collaborating machincry prudU<.:cr). ami fccd lrials uslng roo! arKI leaf silagc and lTII'al are showing pronllsmg resu hs. This Ie<:hnology is appropriale fOl' arcas with a lung tlry ~a!\On " 'hoere lhe availablli ty of fOl1lge is a limi ting f....,l....- in milt produclion. CII"-':H'II ~Ial'\"h. There has bcen recenl imereSl 111 Ihe establ ishme nl of a cass~va stan:h foclory In Tanzania. Tlns in lcn:st has come about due lo lhe increasmg dcmarKI worldwide for nali"e ando in pal1icuhtr. mooificd starch. dTp Studies toc. l repons annual growth rates In demand of 4 .2% fOf Asia. 3.4% for Loltin Amerie a and 2.3% fOl' Africa. EaSt and Soulhcm Afnean eounlnes Impon annually USS 3.2 million uf glU<.V5C and de.\lrin. ",ilh Kenya spending 74 % uf Ihis lotaL [n particular, une indu~lnal emcrpnsc: in Tanzanm. Mohammcd Emc:rpnsc:5 Tanzania Limited (METL). une uf Ihe large~l agro. production and proceS!ilng companies in lhe counlry. ha!! inilialed a di\'crsificalion prugram frum si~al. bases.ova Marl cl Sludy: HU"ncb "1'P''''un,"es for ,~ UM: of ~USl"'. [FAD. FAO. e IAT. e lRAD. l IT A.l\l U . • "-- A Compannvc A ... ly~ .. 01' "'" ~hllelJl., of e ........ oord S"'CCI PtUI<) In SoUI""," Af.-.ca T"" aK of Mabwi. Malft Rq>c)f1 ¡Dn.Ii). Mar~h!ro) 12 Cass a va nnu/" fOI" the food Industry. One of Ihe major biscui¡ manufOClun:rs HI Malaw, (Universal lnduslries). is currenlly employing Icvels of ~round 20% cai>5ava nour in Ih-eir formulaúon. Use is hinden:d by poor quaJily prodoct (co lour ~nd inseel damage). II is eslimated that potenlial demand by Ihis IndUsITy 's 500 tlrear. Thc pricc of ",,,J;,,J;,, (local name ror dry cassava pieces) is US$ 11 1-[)3/1. which comp;¡res favourably ""Ih lhe po ce of whe:u nour. In Ihe inlerviews conductoo with Uni"enal Industries in October 2001, lbey reported a price fOl" cassava "our of USS 14{llt, whC"re..s lhe price for wbeal nour "'lIS US.'5 440. TIIC induMry rcquires Ihe U!oC of swcel casSllva vanelies Iha1 compele (01" de mand fmm the fresh man:et. 1llls suggeslS thal areas of prodUCl,on should be cho.O\Cn far 3"'ay from lhe majar urban marklS Ihal an: demanding fresh Cai>5H'·H. The ,otroducllon of improved chlpping :md drrlOg Icchnology hu sho"'n Ihal c hips 01" lhe desired qLlalily can be produced:11 a compel; I' .-e price (appro~imalel y USS 9t)..100l1), bLll cost eslim3leS oce.:! lo be rLlnhcr n:li noo. An imponanl oovanlage of ¡hc use o( goad qualily chips inSlead of makflkfl ch,ps is Ihal usually Ihcsc chips oeed lO be SCrllpped before mil"ng 10 improve lheir appe3fV1oCe:md qua lily Ihus increasing lile fin~1 COSI. Piloc \esliog of Ih is ICChllOlogy with farmc:rs' groups is L1nderway and wi ll requ,I"C cominued suppon (Icchnical, business and markel developmenl skdls) lo ensUTe Ihal Ihe grwps are capable uf sustalOing lile prodoctioo o f quali ly nOUf fOf indu¡;¡ry. This experience wil! also provilk a focal poinl foc dc:momlr:uing and disscminaling Ihe lechnology 10 otller par1S of ¡he coorll ry, ",bere local dc: mand for higf1 qua!ily nour may be identifi~d. T II ble J. List of companle!ll contaCled in l\1a la 11'1. Lilongw. Caso;&vlI nour fq [ noporGOd I ndu~lri"". Cai>5uva tlour has been SlK:C~rully used io papcnnakiog ~nd plywood manufaclure Huwev.r, the pOI.enlial tlemand for Ca.\!I.3V3 lO Ihesc ,"duslric~ i5 very small. in lhe regiO!> of 100-20(1 tlrear. One foc lory producing plywood (lT1.) has stoppcd L1sing ... heal nour Itnd is repl3Cmg 11 wuh ca.~~"·a nour ror lhe preparal;on of lbe gluc. Ho"cver. 100al demand is JUSI 50 t per year. The cconomic tmpocl is very significan! constdrnng Ihal prices (01' wlleal nour were around 32 kwachas pn kg whcrcas ¡he cassava noor could be purchased al aTOLlnd 15·20 I.:wacha.~ per tg (prtce5 of Ü(tohcr, 20(2). Il Ca~s.anl c hi p!! a mi nour for Ihe dllln IInd poultry (rtd S4:c lor. Maiu: availablllty ror animal feed fluctuates widely in Malawi. During years or drought. mnizc priccs soar and lhtre is a critical netu fOT idenlifying an ~hemalivt energy SOUl\'e. Ca¡sava coulu become (lnc of lhese altematives. Substitution of maizc in animal (ecSlIva chips 10 Soulh "'frica in 2002. Afler 11 sc lhng pnce liad becn agrced upon, (IlS$ 9011. FOB 81~nlyTe). the deal was clllled off bcc.use o( dlfficulllcs in purchasing rne cassuva chips. Thc company wanled 1hc prodUCI deh"cred 10 lown. 1r.lOsponation costs are e~cessi" ely expcnsh'e in MaJawi (5-6 kwachas fkmIt). and the final pncc of lhe chips was 100 high 10 allow fOl" aport. In conclusoon. Ih,s markcI rOl" caHava will becomc: a reality in Malawi, only I( lhe maLZC price.\ Ila:omc "ery hlgh 01' cusava roots can be produced al a lower prieto On lhe other hand. 3.'i reponed in lhe CIISC of TllIIlani3, lhe 1L<;e of ca.~S3va and S"'CCI potalo in \'ertlcaUy integraled d/my proctUCl100 schemcs is also look.ing extremel)" promising. The s;ze of lhe demand in lhis S«IOI" has not becn delermined Ca~~va ~Iardl . ln 1991 lhe m:tri;:et SIU for non·food uses of slarch was afound 780 U)'Cul"'. This dc:malld v.·as mel by importing maiu: starch al USS 6501t fmm Zlmhallwe and Soutll Afnea. lbc $CCIOI"$ reponing the eonsurnption of sta:rch ;m: the packaging mdU!llry. Cíonon manufacture, dry cell manufacture. plywood ~nd textiles. This volullIe of Impon.~ wOlJld nO! just,f)' Ihe establishment uf ~ largc·scule highly mechanised cassava eJ\ lr,¡¡,:lion plant. In local circum,tances, il might be wonh cOnlcmplating ~ smull·scale scmi·rnechaniscd und scmi-tonlinuous process !;upplying spcci fic industries, similar 10 tlle enterpr1Ses contemplatcd for h'gh quality cassavH flour production (see abo\"c) . b) Reyittw o/ ¡lit' dm[r alld filial COl/mr)" wb· .• t:ctor marker srudil's The druft eount ry documents of lhe sludies were recci ,'cd in Apri12002 anu these were re"iewed and commcnl~ retllrncd tn the aulhol"$ ror incorponl1lo ll of correctiuns. During Ihe Sleenng Comrniuee Meellng (Pretoria. 2002), OORle of the main meSSilges of thcse 51udles were prescmed and dl5CU.~sed. 'U ... vn>aJ t..:luOlneO. p.:r1ON1 etlmmunoClUon. 2002 • hboan. E 1995. C ..... VI In ttoo: Mili." OCOllOm)' "'eo.enluto are over !IOO.OOO! and under 200,000 l. O.'erall, T¡Ullani .. has more favuur,¡ble climal ic eondil;ons ando compare tr'llol1cd in Ihis form. In Malaw;, fresh cassava ap~ars 10 be a gruwmg fürm of consumption. especially in lile uman arcas. In the ca'ie of cassav~. l!lere are ooly a fe"" small'scalc drying and mlihng entcrprises in each oountry. There ha"e hecn al1cmpts al lhe mdustri&\islllion uf ~ass~Va as a raw m~leñal for slarch c:\lr.il:tion and animal feed in Tan1.ania bul these have not prospered. dile 10 lock of raw m:llcnal and low CO~I of altemalivc produClS. Mechanised ind ustñal procl:ssing of cassava and sweelpolalO for fnod. fecd or Olher purposes is non·exislent in Tanzania. Mala"'; and Zambia. Supply chaios sludied The sludy e~aJlllOed in detail lhe followlllg cassava and sweet potato sU[lflly c hains. 16 In Tanz¡Jllw , 1. fur sweetpotatn a. Fresh sw~elpoMo supply 10 Dar ~s Salaam fmm lhe ,(\;ISlal r~gion . h. Fresh sw~elpot~lo supply Iol Dar es Sa!aam frum Ih~ ,entra! region. c. Fresh s""eeLpotalo suppl y 10 dislriCI ~nd regiona l ,nl)l lals fmm celllnd "~OO d. Local supply of eassava and sweelpotalo (frum village 10 vil13ge) in Ihe cenlral regioo. 2. For raSS3 \'3 a. f'resh caSi3va supply 10 Dar es Salaam from Ihe cooslul region. h . Fresh cassava supply 10 Dar e.~ Suluam fmm lhe central region . c. Fre5h eassava supply Iu di)tricI ~nd regional capilal~ from lhe cenlral reS'0n d , Local supply of fresh cassaV:l (fmm village lo village) In Ihe cenlral reglon. In ÜJmbu.: 1. Fnr ~weet l)Q l at o: a. Fresh $wcetpotato supply 10 Kilwe (copper bclt) from Solwez; n. Fresh swcetpotato suppl~' 10 Losaka frum Solwez; c. Local sur]>ly of fresh sweetpotaw in Solwczi 2. fo"o rdrl~-d CIISSIIVIl ch ips a. Dried C:I$Sava chi ps and nour supply 10 K,l",e (copper beh) frum Mansa h. Dncd cassava chips and nour supply 10 Lusaka from Mansa c. Local supply of ca,sava chips in Mansa In MaID,.,;: 1. For swet:lll'Ola lo: a. l'n::sh swcetpolato frum Zomba di slricl lo Zomha and MalosaINttmwcrn m:.rkelS h. Fresh swcelpol:'lo frum Lilomgwe dislricllo Chimbiy:.. Miluru[u. KawaJc3 and uloogwc markelS c. Fresh S""celpo::Mato fmm Mzimba dislncl lO Nkhata Bay. Mzuzu and Mlimba Boma markels 2. Fu r alSSlIVQ; 3- Fresh cassava from MulanJc dislrict 10 Zomlm and MalosalNamwcr .. m~rkc! s b. Fre~h cassa~a from Dedza di~tricltO Chtmhiya. Mlllln e:!u , Kawalea ane:! Lilongwc rnurkelS C. Fresh C35Savu from Nkhala Bay distric! 10 Chimblya. Mltundu. Kawalea and Lllongwc markelS 17 1lIe mform:llion was aggrcgatcd ond presenu:d IR a way th¡U it is Illmeult tu cstabli5h the compctlti\'cness of eaeh of thc.sc chains, and compare belwcen them. Shoold any fmure projecl scck lO make inlervcnllons 10 Impl'Ovc the comreti th'cness of any of 1he.'IC supply cnams, Ihe data generatcd could be furthcr anaJyzed and disaggregaled by each speciric set of OCtnrs in lhe chain. of imeresl. Market dema nd Cilies: Fresh ",.SS!!"!! IIDd s weelpolalo. Tradcrs holh in TlInl.nnio ond Zambia reported no constraims 10 dcmand for fresh rool~. Cunsumcrs onl)' noted 13d of 3vailabihly of swcclpolaro during ccnain times of the year. In Tanzama. farmcrs report iocrc=d !;;I.les uf cassav~ and swcet potato In both producíng region. swdied. 'lñis suggcsts that the dcmand li increaslIlg. In MaJawi, a quanlitative estimate of the dal ly volumcs of fre~h cass.ava and ~weetpotalO cntering Litongwe eity Wa$ m3dc: in the 2000.2001 season_ In e.\cess uf 40 tld~y of cassava and 30 tlday of swcetpot.alO "'ere rccordcd. A similar stody undcnaken in the following 2001·2002 season rOl"" c;m;ava 5howcd hlgher "olumes tIlar. might sugge..~1 an mcrcase In dcmand for Ihe frcsh prodUCI. Cilies: Dried chip and nour nmrk t t in Kilw~ (Copperbeltl IIml (.usakll . The number of tradcrs of CassuV3 chips ond noor has iocreased, anu fanners report inereased sales over past 10 ycars. Tradcrs reported inereased volume •. This SllggestS thal the dem~nu fOl" dncd chips .md nour i, incn:8sing. In Mala ... i, 89% of households reportcd processing cassava into some: form or othe,., Ineludlng ur)' ~hips ur makaka. However, tradc in dry cassav~ is muinly conccntrtllcd In the North and Soulh of the countr)'. Intermedillte lown~ . Then: is conSiderable trading uf swcctpotato o\'er shon d,staoces m the ceninll n:gion of Tanzanla. mdieating local dcmand by pcrsons that do nO! gro .... the crop. EX I)Or1: S .... t.'l'tpotato a nd e&(!\ava . Thc studies did no upticlIly atklrcss expon of swa:tpotato and cauava, Ho .... cver, j¡ is t.:nown tIlat 10 lhe case of swcctpotato there ;5 infonnaJ ~ border lradc. wuh regionaltruUe opportunities from Zambl3 10 BOI~"'·ana. Namibia aOO the Dcmocrallc Repuhhc of Congo. ror cass.ava. Ihcre is onfunnal cross hor&:r lrade of chip. to the Dcm\.II:ratic Republic of Congo from bmbill " Consumer hlolblts loInd II rt'rcnonces F~h swt't'lpolalo. SweelpntalO '5 Slill predom .... nfll ly ¡I food o r the low-.... comc CQn~mer group, but i$ apprecioled by higher .... comc e~m .... g gmups. For example. SWl:etpolillo 15 a\'allable In supennarkels in Dar C~ S~l:1ll1T1. where consume!> prder swcctpolato room th.e cenl",l region. S"'cct poIato fmm Ihe coosllaslC$ sallY. The pnce of freslt swcctpot¡llo is double or tnple pnce o f fre.~h ca.l..'ava_ ""ith ~ marked season~lily . and Jargef price tluCtuallOOS lh~n fresh cassava. S"'CCI polalo lB mainly consumcd as breal.fast replacing bread ~nd O1her wheat-bas.cd products. It is con.itlered less ~el>atilit y CQmpan::d wilh cassava, w¡th hm1led proce.\sing and product/n:cipc ~pp[ic~lions. Additionally. In sorne .tores of a supcrmarkct chain recenlly opcned in Malaw; and Tanzania. il is common !O find frozeo sweet potalo sliccs. coming frum Soulh Afnca. Fre~h taSS,II1"II_ Cas.sa\& IS al$O ~till pn:dominW11Iy ~ food of lhe low-Income consumer group and IS the cheapesl carbohydnllc !>OUTCC aV3Jlable. 11 is \'crsaule, ,,' ith ",ide: v;uiet~· of diffcrenl fonns of prep:lrahon. belng used both lI!i a ~Iaple. oomplemenl ing ug"¡;, rlC(' etc, and as dchcacy rOl' cenain trudi tional festivals (e.g. Ramadan). In Dar es Sal~am, il is increa."ngly bemg sold by (ood ,'endors/l1awl er5 un lhe Mreel as a snock roasted, along wilh roasted green marze. Dril'd C3'!S11\'a chl¡n; IInd nO!l r, Mll11ng mduslry wurk~rs 111 lhc Zambian ~opper bel! come fn)m Ca5Sl1Va producing arcas, " 'here 1m:) are acclIslol1led lO using dry C3S$3va chips Ul' !he bas is of fOO!! prcp3rJlion. High maize pnces in Zambl3 have meum J shih from !he use of mai ze lO dry CIISs:\VIl in the productlon of uÚma. This also QCcurs 11\ Mala",i in yean; when lhere IS a ~ honage of maize. In Zambla therc has bcell ¡:n)wlh in the 1Is.c of dry chips as J snack sold roasu~d by streel \'clldors in lile copper bell. occompamed by groundnuls. Tradlng aOO lranspvrt There has been I subs!anl1 31 mcrcase 111 lhe slI1all lradcr c1~s.; as a h ve1ihood OC\lvlly , and supply chalAS rrom flffITlC'!> 10 COMumers in eilies lnvo!ve belwecn 3 lO 8 IICtOrs, Glllding of fresh sweelpo1ato and c3Ssava is an importan t value addin¡: ocliv;!y undel1aken by 1I1Ide:n;. In general farmen; do Ilot grade before sale . c.>;cepl in cenaln cases In Zambia, 1ñe major price fllClOrs In 311 counlries are Ihe sile 01" roolS and vulume. Neilhcr fresh ~weclpolJIO nor C3Ssav¡¡ are bou¡:hl Of lold by we igh! al any polnl along lhe supply chuin. Truders of fresh produce buy Ihe erop in fiehJ. organize harveshng. haggi ng nnd Illlnsponallon 10 Ihe markets (e,g. Tanzania), Transpon by ruad tS Ihe only nIC:I1I1S of lran~femng Ihe produce from rural 10 urlxln arellS, and lransport C0515 are [he h.ghesl proponioo of al! off-farm cmls. SUI11lly ot cauana Qnd swet'lpolDlo In genera l, area unde:r cassava aOO S"'eclpotaIO is lcss !hen 2 ha per rarmer. In Tanzama, (resh CIl$S3V3 and SWttlpo!alO fanne!> produce for both home consump!ion und uk. Amo ngst s"eetpota!o farmc rs thefe arc a numbcr !hal are more commerch.lJy onenled. employing commerdal procticcs (monl'l-Croppmg, ~illgle varielles. seed Iled~ 19 etc.). In all counlries. Ihe mllf"ket orienlalÍon o( f:trTnen; 15 grc.alcr Ihe closer they are lu major roads and urt>an cenlers. TlIe impl ication of mis is Ihl1l bwlwng mads willlmpnwe farmer lICeess 10 rnarlia. eass.vII ratmen have accommod1l1ed Iheir prodllCllon prac1Jces lO pn.lI.h.Jl.:c alllhe year round. Howevcr in all cOU111nt$. sweelpolalo prodUClion i.\ ~easonal. The slUdlcs showcd Ihal fanncr.l are producing lhose vaneties mOSI prcfemd by traders and consumcrs. T1Icre 15 00 use of porcl'la~d InpulS. CJIeepl In ver¡ sclCClcU cases and by Inrne more: eommcn:ially orlenlcU farmen (moslly planlmg malcrial: e.g. swt<:lpOIalO vmes). F;umcn seU IMl r produce in lhe ground. I'Iarve)ling 15 unckrtaken by Ihe tladcn Ihal purchase Ihe trup. TIte ¡mee: 15 e~IImaled by ficld Inspccllon by the Intder ond bascd 00 planl age. Yoriely. heallh Slalu5, ele. The slUdie~ cnuld nOI dclCCI aoy innuen~e on producllon pr1lCliees and relall"e accesii to markcl (e~eepllhose ca"", menlioned aoove), Swt<:lpotalO has OlO'ed fmm Ix'iog a largely on-fllrm home coosumplion crop lo:a eornmcrclol erop. aod as a conse.:¡ucoce men havc bccnmt more: mvolved 10 produo,;lIon and InIdo: . Beforeh:md. S\O.telpOhdO producl ion was pril'lClpally Ihe domatn of womcn. As CJlpccled. farm gale "nces ore higher closc:r 10 urban markels. Thls \O.'uuld pn:cludr: Ihe pmcr:ssing of Ihese Crop5 c lose 10 uman Cc111ers. bUI provides 3.n opponunlty for lhose areas dlSlall1 (rom uman ccnle~ \Ir'ilh poor mar~el aeccss. In all countrics. Ic~~ lhan 10% of farmers partieipatc io assodalions and Iheir !lCcess 10 techmcal assislJlTlce is p''''' A mInI won..shop"'as hcld 10 Nalrobi in Novemllcr 2002 lO finllJize Ihe ~lUdles and pre~ synl heslS documcnts th~1 pullcd logelhcr infonTIlllion from lhe I:omprchensive liternlure review. qualilmive prc-survey aiM:ssmcnts and Ihe quan li lali ve sub \.Celor slud y. O\"l'rllll coodllslon.~ l. The slodlcs gener.llcd an imlTll.'nse IIIT1Ot1I11 of informallon. The wmmary documento :allhough n 1$ comprehens, \"c and covcrs Ihe muin eomponenls un whlch IOfomumon was g;tlhered. 15 nOl easy 10 digcsl. lt ~'Wl¡J have bcen lTIOfe iIIuslral ive ir it had madr: grcaler use of Il,blc$ and figures 10 compare inrorrnation IICros~ counlrie!>. An e~amplc is Ihe cnmparison of markel pnces and produclion CO~I. In lhls regard. il was unfortunalC IhuI 111 Ihe meeting held in Nairobi in Novemhcr 2002. 10 bring al! lhe inforrnalion logelher. Ihe sludy lcadcr \O.as unable lO allend. In relrospecl. more tu1lC eould have bcl:n sel aslde fm analyzlng 1M data once 11 hOO been lablllated. Wh:1I Ihls means is th:u signlflCDnl fuMer uOIII)sis could be undcrtaken 00 !he dula galhcred. ",hith cOllld Ix' useful ror any fllnhcr sllIdy of S .... ·CC'lpOMIO ur Ca5sava 10 Ihc respeclive coumn es. 11 would Ix' hoped Ihal lhc basic dala could be madi: availab!e. $hould other researche~ requlre ,1. 2. In 311 [hree eOlmtries. eassava as a holtllUl fooIeTe 10 recngmsc Ihis as imponanl. nddi lional und complerncnlary skllls lO its e:c: of cassava 10 ammal fcediog. loillally. 1",'0 smaJl·scaJe eApcrimenls "'ere cooducted in poultry f.Lr es Salanm \'.·llh lhe malO obJe(:li ~es of geniog SARRNET slaff rarnil iarizc:d ",nh lhe use uf !he soflware used rOl' furmulalJOn or Ihe fced as well as Iu gllin sorne praclicat e~perience on lhe conduction of blological e~pcnmenls, The lay OUl for IWO cllpcrirmnlS, one for broilers und one for layer.; ",as descrihed prevJOusly·. The imual Idca ",as 10 ev~ J lI~1e lhc effect uf a pantal repbcemcnt of Ihe maize wilh cas~~"a flour Icvcls uf 25 und 50%. Perform~ncc 01 !he anim~ls "'as mcasured againsl IWU ~ommereial diet,~. The seale of lhe expcrimcnt was • CLAyUC .... e t ... T. SAlUl.NIIT. lIT .... Uf' :!OO:!. SUb<1mlncl ",~_<)I ro. the u«IfIJOfI " r us ... nl/SADCJSAlUlJ\'ET Gnni 1'00 690-G·OO·99.(l()258-OO bo:l~n ,11$ J¡ ......... """"l 1...u'ule fOt T,,,p'''''! ...."...wl~ . lITA. and !he l"';periment proeeeded. 50IIle mortali ty, sevtIl: in sorne cut!., occun-cd. This monalily was causer .... 1t" .t pub! .... p;vtncfS ..... J BuiIJqo SARlUo"'ET telm B.O.pin. -:s..n. ..-.c.. ....... hove oI..-ody ....... "'""" •• • .. I.obk .. SARR."'El· a llice> 11 s...m. do1.. oJ..->d, tocco. modo ..... ,0lil. al SARRXEl" oIT....,. <· .. so ... ( ¡n""'ed) TecMICII dlla "'.,llb~ l:c<> .... ", i< an.I )'>" o( f~. ulls ellbocltcd (pcooin¡) I Ec:ooom;c; dm ...... Iabic '" In 'm,1 feedi ng·· ' ransl.tcd inm En¡hsh l oo ,vliloble fe ¡>I1oI pro,cc1 l'(>Wrr ",,"" "..=nl.,i"n . inpoulu~ " Ffllsibilily sludy on Ihe use or =~'illnl In IIlIlmlll rete! in TanUlnhl , with special fmphasls on pooltr)' reed _ A summarized reportU Ovrrllll objf'tlh·e The overall objeclivc of Ihis Slud)· w;¡~ 10 evaluate Ihe feaslhi¡'ty of Ihe use of cassava as an altemulivc animal ingrediem 10 m:lIze in balanceo feeds in T~nzania. Due 10 Ihe Uommalll ro l ~ o f Ihe use: of animal fecd ror poullry prOOuction In Tan~ania. lile worX cmpha~ized brolicrs and layers. Feeds b;¡se:d 0fI a comhin:l.1ion of C:LSsava aOO 5OybellJlS, as an allernall ve ing.rc:d ienl 10 fishmeal. wen:: :lIso evalualed. Sp«ifie objeelh·es, I To analyze and document lhe roo! am.! leu'·ClI pruducuonlul1hzllllonl.I)~lcm fOT dlfferenl pruductslcnd U.>et"$ (co~ling. yields, quahly, labor implü:atiuns) 2. Tn e,·~lu,de Ihe pcñorman¡;e uf broiJcrs and laycrs. fed with ailern~lI vc leed, preparcd wilh cassavu products. usmg f~co furmuhllion software. 3. To analyle Ihe lK:ccplnnce nf fnrrners and fccd millcrs of cassavu as animal fecd 4. To make rccommcndatiuns for future researeh IInd developmcnl acti vitie~ fur pooltry and dairy 5CCIOI"S In Tan;wnia F«d lrial, Dala 00 \he currenl animal fccd pr.lCtice~ in TBnl.llOia was obumed throoglt review of lhe IlIenHure avai lable IInd o speclfic ¡urve)' dcsigncLlto OOlllin informmion from ke)· sllIkeholders in the animal feed IInd Jives tock seclOr. A lotal of II differenl ~umpanic~ were imervicwed including feOO mi11ers (4), poultry farmcrs (4). and dairy farmers (3). Table 4 prese:nlS a SynthCS1$ of Ih-e infonnmion given by key 5takchol~rs 1Olervtewed. Informaciol1 obtaineo was use:d as an mput for Ihe formulauoo of Ihe CJlperimc: nts . Two e~penmenlS \I.·ere cOflducted: one wllh bruilers IInd one Wlth la)'el'1. The expenrncou were limited 10 !he poultry .sector bc:cause of the low invesunenl and rnaintenan.ce CO$I$. The foUowing e;eclion presenls !he dlSCLlssion of Ihe results obtained aod wme preliminary concl uslons. Discussion As mentioned befnre. Ihe survey wlth poultry sector st:lkellolders was conduelC{] "'·l1h lhe objeellve ofilelping to Iden tlf)' opponuni tics aOO OOlllenecks fOf a ""dc:r Llse or CassaVH in tile ammal fecd SCCIor in Tll/ll.llf1ill. E\·en hefore conduct iog lile inlervlews, SARRNET liad II$sumed thm one Important SUlU:SS factO!" of C8$SIlVD. lIS an altcmal1ve rttd mgredienl was ilS repulauon In Ihe animal feal SCt:IO<". 1"herefore, Ihe fecd lriuls 10 lhi$ sludy were not enl)' coodocle<.llu oblllm ,nfonnmion 00 liS nUl1;liunaJ va lue. which has heen demonsll1lled in numerous previous tnals. bUI also lO demonSlrute Ihe potcmial of cassa'·a lO Ihe an imal fecd sectur (funncrs, fecdmillers. consumcrs uf animal prodOCI.~J " DeGrool. Ddlby Thesis """k Tropical " ...... 1 H~y 79O-101·2~-OJ(). W.",",n¡en Un,,,,",l y. H<:olland. 2OOJ. Dnf, doxumc.". SARRNET. Tlnt.a n,a lOOJ Tahle 4. Opinions or reedmille r.~ nlld poullry farlllHs in Tnnlanlll - A sur.-ey. ObJc<:uoru; .",'M e .... " in .n"".1 feed • • • • "1II:un";ncG ,lit Umc rr. fea, "rloos 'nl dieoo • Anlmll prodl.,.rs ;oJso mmpllln IbQI¡l 111lCILI:lUnll ¡>rice;¡ of raw mll"",.I, • • • • • • • • 0\11<1 feedmllen reir on Opoe ,..,rh, n rhe f'lIIl ly Top rommorr;aI form ha< .,:,:c,.. \o rhe ad' ''$ of. OClen,,,, fmm Nairobi " "r.",:" ... ",:' (,nd c" .. ". roo! mc~1 (CIUoI ) LI.) dU>I~.,..¡ """Id nccd" pelleuoed !O be Ible lO i""hde n In ,he .nlm.1 feed. A pellelile, ,.-,11 he ''''1uored .1Id fIKISl fw:hnill"",.,..., f'IOI ""Ihna !O inve<.! In ~h,,'lCt)' Oneol ,he reedmllen, did UoeCaoN'" roo! mell fo •• pcrKod uf." momhl In brooler and "" feed (6!n CRM 1M 1", .... 'ul RmolCl' performance lO'" lhe ume 1M pi,pcrro. ........ 'C , .... 'IUJCd "ml)'_ AcCOl'dlnl lO !he leedrruler. tht ioch .... ., .. uf z,,,,, R. nhetlC1"" :H • , ,~ Table 4. Opininns or feedmiLLers and pouLtry rarmen io TllnUlohl - A 5une}' (Cunlinulllion) ===r~~;;.~~.,~~~ , .. ".~ dty>nM ~ ...... kc rttdmlne" ar .. id ofus1nl CRM I'I.:omin,.,......va ,_ loJ1J laborlW!o ObjetO""" ."', ........ "", .• in .... mol feed le"""......",) • • • • • • • • • • • Fecdmikrs uc dcmandinll th>. casuva shouJd be duppoJ tal ¡:nfcnl>l)·. I'"llculoCd; cauava r ......... " ',llllOl be "bIc 11) .... k. any ,nvc..lmcnl. Fumen. noed credil and ."""",. '" """'''' lb-loOtlalMIII> Maid n:a.un Iuo- ,..,. prud""lD~ .,...., ... ~"'''' i. !ha! "" ... ...... va ......,. .... f:UOIl and ", .. ,:h mor. pmr. ... hk. <1>J1 uf (,,:.11 ~n.\ll'" in 0.. a Sotum, ",,,,1"1 ta,,,,,,,,, ...... rrWzc brIn, ",h....., '~ch<.:l"",. So,.,.., l"",t"Y f.nncn """"de. casu'" fOOU.' In ,nfenor prodl>.'1 and ... 11 ....... 10 ...... ide""" of!he. roalrary bcfore !he.y ronsidc. I ch"'t< ,n thtor feed formulas . ."'~ Many.t.iry f."nero "'" C""""YO. ~= .. o 1""' .... '" ro.. "m IUr~ ..... ng ~ __ .a le .. "" .. c""lC<.n " .. Iy CUI th< call. oldu ¡nSl Inec.use lt l. ~dJmcult 10 be Iwvesto:Ilnd ron ... ",... more ,,"'" Fob$c,,·ed on """fU • ., in animal I'",formanct 1M , • M,,,, "." " , " • • , , · "", • " , , , wh •• l lhal t< 00 longtr "",fu i fo. human " ....., -",}b :an it i, 1"0 . ' I'"llOiv<, aM ltOI .I .... y. 10 be: ,mported frorn Ind i. or Brazol'M il ni",,, ------------~. ~ Cor"hlioru fttdm,lk,.. "o", 10 .... ful(¡llrd before """11 <;o».va SAkkNI:T • • • • • r "",n;,,,, ilil A l.bo .. III'~y 1<11 Of any (1111<. proo( of Ih< nU";lIo ... 1 ,-al"" of!he caosa'·. ', .. mula ~ 'IIC'CU\r:¡eed m.1 , · , " . ..i, Adopllon·polcntial rQr cas.~a \'a in Ihe un lnull reffi. sector In Tanz:mia • During ¡hc IJIlCrvlCWS wllh fcednullers il w;c¡ mentioncd that lhe ehca!wg wuh rJW malerials. by iIS 5upplicrl. Inhiblled tltem 10 reach the nutri¡iunal vHlue indicaled hy Ihelf (eed fonnlllll. 'The)' suggeslcd supel"'o'ision oc qualily control on Ihe ~uppH~ uf ro\\.' matcrial in order 10 be sure uf ils nutrilionaJ valIK: . In Ihe case of cassav¡t. lhe leaf !.lu~Jily vanes grently .... ith ¡he ~gc 111 harvCSling (HCN). harvcsling u:chniques (1eaf- Slem ratIO I digcstihilll)') and processing lcehmqucs (HeN, Slorage·abilil)'. palalal:1ility). Chcaling by supplicrs wilh bad qUllhly cassava. low in nUlrients and neh in tannins. could senolls ly damage ils n:putalion. For Ihis rea5Ot1, iIS produclion mU51 be supcrvised Qf Ihe final ingredJCrlI musl be tesled IInd cJas~lfied inlo ~ qUlllily- calegory before bemg supplied tu lhe (cedmillers. • • • • Most o( tite ($malJ·scaJe. lo .... prodllcti"ily) fecdmillers f1':ly un their O"'n and ¡helr (amUies' expericno;es in feed formulalion 1100 do 00l use u murilionist or a compUI~r. Usuall)' [hose f¡;c,dmi llcrs prefcr 10 stick by Ihelr traditional formula and ~re 00l ver)' pen:eplive 10 the IdeO of using CI\SS,1Iva 8S nn altcmative ingredienl. Thcy could he inflIK:oced. howevcr. by a goad repu[atlon of casuva amongSl other sllCccssful. feednullcl1. It sccms logica! 10 conviocc tite smnl1 groop of percepUvc. top m3l;';et · oriented animal feedmlllen. 10 expenmcnl Wiln ~assava. FcedmLl1ers cnuld ~ con vino::cd when Ihe potenlial profil g:lln ..... ns from Ihe polcnual risk. ll1C' poIential profil gaín could be demon$lJ'3ted with ongolnS brollcr aoo l:Iye r tri~Js. Whell Ihe f1':SUllS uf fecdmillel'!l, wno ~dopt cassava based reeO:l are goad. iu rcputauOI'1 will ino::reasc:. and Olhcr fecdrnillcrs mlghl bt lempled lO follow Ihe eumple. Musl of tite broiler and IlIyer farmeT'll . whu utiJi5C up lO 80% of lhe domestic commercial feed prod\ll;ed., purcha5C lhelr animal fcod from fecdnullcrs wilnoul I:nowing lIS (prttise) nutritional conlcn!. Assuming Ihat brOller perfOlmance wLlI nOI tlrap as a resu ll and gai n a bad rcpolollon. the C¡lu~,·a·based :lmmal fecd wiJl ha"c 3n immedialc market ooce il is accepted and produced by lhe feedmillcrs. Oairy ,,"d pig farmers usually make !he;r O)wn fecd fonnul~lion. purchaselgrow [helT own ingredienls IInd mlx lhe ir own fced. Spreadinl!; !he reput~lion of ca\..\Ilvn as ~ potcntial ~llema1Jvc animal feed ingrcdicnt 10 maile IIIlghl be more time consumíng ror Ihis group. Oairy and pig farmeN IIppeared 10 be le" wtlling 10 Cllpenment wilh lhelr fced fonnulation ¡han [he feedmi llers. nod could necd more dccl$lon.time before adoptmg coss~va Some feedmillers find Ihe cassava·hased recd [00 dUSI)' und insisl !hal 1I needs to be pellelized beforc U~lng;1 in their animal fecd In arder lO mal:.c cassnv3. 1001 chips and 10 pdlelizc [hc cau~va bascd an;mnl fccd a chipper and pellclller are rcquircd. Feedmillers :ue nm willing 10 Inve~1 in lIew mochinery and shari ng a pelletizer or chlppcr wilhin a gruup of fccdmlllcn could pruve problem~lIc doe 10 Iranspon tus'" uf !he fecd :md communi~ation tliffi cu lue.1 011 taplCS such as time schedule!; and mllmlcnance. " • • • • • • - ----------- An a1u:mauv( could be 10 pro"Id!:!he fe«lmlllers :mdlOf fa~rs lOo'ith lhe: possibility 10 hin:: Iheir pel'SOtlol chipper andfoc pt'lIetilcr. They woold have ¡he mnehinery 00 Iheir own dOlTUlin. diminatlng extra transpon. cost¡; or dela)' in production. Thl: di s¡Kh-anlagc of lhl~ ,<;o luhoo i5 ¡hUI lhe producllon costs of cassava diclS \\Iuuld tncrease with ¡he reol COSIS of mochlncry. This means ¡hui wheo lhe rcnl is high. producing cassa"a dlcts will only be possible for highly producllve feedml11s. The fixed cOlas of¡he mochinery can Ihe" be dividcd oyer mass prodUf.'tion. AOOIher problem 15 ¡he unavaJlabdily uf ~·lL.'ls:.ya for Wlim.:1i feal. In :lnd nmund Dar es SlIllIam and olner mcdium 10 large 10 .... oS, cass.ava 1& producW mainly fOf human eoosumption. 1ñe price of frc:s.h cassavlI I~ omund 90 TSh. Funher away from IOWns and manels CIISSlva is Ilvailable fUf 11 much lower priet, whieh can he :1$ 10w as 15 TSh. MOSI of lhe rela li vely high prooucuvc fee¡jmillcrs in Dar c~ Sulaum own Iheir own vehiclc(s) for Ihe Iransport of feed (ingredicnts) . When Ihis group uf feedmiller.> will make an agrcerrn:nt wim cassava farmcrs. who are 100 far awuy from mW"kcIs 10 ¡¡ell Ihelr fresh CUSSllva for human consumpliun. Ihcy .... ·ill be able lO produce CII.~!o.1V3 basc:d diclS 3gainsl much lowcr produ.c lion CO$I5. Asid!: from h3\"1O& CIISsava ava.lable lO thcm. feedmillen could seU lhe.T excess cnssava 10 the smaUcr fec-dm,llcrs aod animal farmcrs. against a small protil bul $1..111 below lhe: price of fresh CUSH'·U. 1be planllng and harvesling nf cassava i5 luborioos and afllmal [armers ..... ill necd 10 empluy more people. usually men duc 10 lough wurk.load. 10 look afler Ihis IUS\;.. High labour COSIS and a reMively low priee for CIISSllva eoold be a rea~on for animal farmen 10 buy lheir cassava from elscwnere. lnstcad of growing cassaVQ Ihcmsel\"cs. In Ihis $Cenarío. feedmlllers and farmers will have IICCCSS lO cheap ca.~s;ava, und car.sava farmen 10 rur,,1 W"ClIli away from lowns ",m have new ¡ ..... "Omc. The clIUlIva farmc:rs ",ha are largeted 10 becomc lhe supplier 10 fecdrrullcrs in DIU" es S~hUlm. should be informed of impm\"ed planllog aod hurvesling strategies io ordt:r lO ensure a uniform quolny of lea ves aod ruoIs amongsl all casuva fanns. The umounl of labour nceded to nm .1 fcedmill will not chnnge with !he inlruduction of co~nvn. The labour rn:ctS.l'ary for weigto ing and nll~ing dQes nOI dlffer smongsl ingredienlS. However. e~ITII luboor 's nceded 10 pelletize Ihe cassa"a-based dlel! . FeedmllJers do no! see thi~ as a relev31l1 prob1em. Gcnder dlVisioll ... in 001 thallge. Ir new emplo)"ffs 1IR' rtecdcd lhey .... in be male: duc ID lhe heavy .... ork of caTT)"III& fced lO and frum thc: pc:1Ie:uzc:r. Sorne dairy farmers are ~Ire:,dy using ca.\.Sava Icaves as 11 ~ubstituee for gran. Hlln'cstiog the !caves is far len labonotl~ and Ihe qu~llIy is beuer. Mectings should be WTanged, JOi OlOg dairy fanners who are us ing cassava and lhose who are o()\. Thi~ way, lhe: good Tl:pulatioll of cassova leaves wil! be spn:ad. Scientisls with kno .... ·lcd¡¡e un gro ..... mg ca\savn. harvcsllng cassava :md posllive efrects of cassava leaves on dairy performance should ¡mcnd lhose meelings, possihly wieh resu lts of a d!:monstratlOn trial. in order 10 help che 'newcorners' wi,h gcuing staned and lhe • • Olllers wilh opIimising Ihcir dai ry perform3ncelprodllC!Íon COSI5, During Ihe mee!lngs, Ihe 1,15(: of cassava rools could be inlroduced. Cassava mol chjp.~ can be us.ed ~s an energ)' $OUrce in laclatmg dair)' ce", ('atioos· 6 weeks) per Irclllmcnl groop. whieh explains the vcry high standard Ikviülion. In OIher words. thc rcsuJts from Ihe rnullivuriutc unulysis are not rcliahle due 10 bias nnd high standard dcviation.~. In Ihe lasl week of Ihe bmiler trial (week 6). Ihe broiJers stalioncd al Mikochem bccamc sick WHh Gumbom discase lUId Mafua hac lena. The first mortaJity cases were for the 5O%CRM lrcalmcnl ¡roop, suggcsling il Iu be !he mO;SI susceplible ¡roop. In a prcvious sludy by Ajan; (2002), thc 5O%CRM trealment group suffered from a 42'1> monalily due lO Iyphoid, wlllt the cause mosl likc ly being lite Safmo",dla I)"pharlllrium IItal CQOtaminaled the: fishmeaL 111;$ was not lile casc here. wllere a higb mortal;ly was nO! confined 10 lhe Cjlssa~a groups. The mortal;ly Tale ranged fmm 19% f(}1" Ihe O%CRM lO 38% for control 2. The rc5uhs from KIgamboni. wilh lem mortal;ly. ~onfirm thal rcplacing matle witlt CRM up tú 50% has no effe.;:t on broiler mortality. Thc statistical annlyscs of lhe rcsults gave nn ,~;gn ificant differen~e for mortnhty helween IreatlncnlS. " ASHlc rmm lhe dlsease oulbrea),;, Ihere mighl be some IO leraction bel\\leen gm~'l h rute und the relocatlOn or lhe chic ),;s in wec ),; 3. glving anOl her e~plan3l;on !O lhe difference in results bcl\\leen bmilers in Mi),;ocheni and Kigamboni. The trealmcnl gmup. where maize \\las fully rep]¡.ced by CRM. sufl"ered fmm diarrhoea 10 I uf ¡he 6 wech 10 Kigambonl Ill1d 6 OUI of 6 wec),;s in Mi),;ochcni, wlllch could have resultelc 10 pelletize Ihe feed. resuhing In a lower feed Intake and a dcprc:SSIOll m growlh rJIC . The dlfference~ for groW\h rales betwecn Ire3\rncnt gmops wen:: si gníricanl :.ccording 10 5IlIIislical analySIS. Even thoogh growlh rotes "'"Cre ¡O",'es!. fOT Ihe tass.¡¡,\'a trealmenl groops, Ihe ~'os t of produciog J kg of brnller meal in Kigumboni wa.' masl favaurable ror thc: 25%CRM ball:h (427,51). The twu controls were IT\(lst cApensive (603.64 an¡J 618.99). Thc r~ding eosls in Mikoclleni were hlgh ror all lrealtnenls, rangillg from 708.91 (25%CR M) 10 1095"~6 (O'l-CRM) dile 10 ootbrcak IIr ¡Jísease In "'"eCk 6 and h'gh wasted rccd frocllons. The:sc r~ull5 arc somcwhut mlsleading t«ause 100al Ol.llput (in bruik:r meal) should also be taken 'nla aceou nL Even though thc prodl.lClion OOSls are decrcasin¡¡ when using Ih.c aJtemuth"e ingredlent ea.U:lVU, lhe broilcrs rellCh a lower .... ·cight. Broilers in Tan1.allia are uSlJally sold al a wc'ghl of 1-1.2 kg. To reach thlS H~erage weight. broilcrs need 10 he kept lInd red ror 6 weeks when reeding Ihem wilh 5O'l> maizc and 5O'l> CRM of Ihe 100ul dl el (25'1<:RM). The 5O%CRM bateh di¡J nOI even reach thc I kg RI lhe end of week 6. Thc control gmt.tJ}' and thc: ~ . (;RM blllCh rcached an average WClghl uf I kg in ,""cck j or c~cn in weck 4 for conlrol 2 In MilochenL This mcan~ \hal in pr&clisc:. the b:l.1c~ fed WI\h maize-based ¡JielS are rcady ror.\llle al ""cck 4 or 5. In olher ",'on.ls. in reahty lhe total fced eoslS for thc maize·bascd diel.$ are lower then Ihe rcsul ls ~how . Still, ir th.c pricc IIf maiu irn;rea.~s in lhe fUIUn: andlor when lhe: feedmillers will pureha.o;e Iheir CQsava rrom diSlan t cassa\"a farmers al very low COSls. Ihe repla<:ement of maize w,lh eassa~a CQuld be eeonllmiral. Pellclizing Ihe feed wiJl. as mcntioned befOTe, inc~ase fced intake and growlh ralc. Comparison or clusan lo maíZ/' ro r laye r r«d IIum uraclurlng TIte upcrimem~1 d<'Slgn of Ihe layer Indl m~o.Ie il impoulble 10 condUCl any statisl iea l anal~I$, sino:e there wa~ only MC lrial or source of data per Irealmenl group. Wilh no repctillve re~ults 11 WILS nOI posslble 10 calculale a standard devialioo or tlle s'gnifie~nce IIf \he fced foctor In Ihis layer lrial the egg productlOn wa~ nCg!lti~ely effe<:ICd by cassava. TIte CMSa~a tualmcnl groups liad a 10""cr egg proIluclIon ¡hen lhc ~ontml group, of 27% for Ihe 5O-no bateh aOO 19% for Ihe 25-no batch. Thi. mlghl be due to lhe facllhal tlle fec¡J was offercd in mash fonn and nOI In pclletcd forrn, llJ; done in lile pre.',ously mcn lloncd ¡a~"Cr lrilll by Smilh (2002). ClISSlIva did not ha~e a SIgnificar\! effe<:t on lhe egg size or egg we'ghl. Combllling so}'bcan ""lIh ca!l5a~1I hall an cvcn ru~r negalÍ\'e dfeet on egg proIluclion, Soybean is more CJó pensive, lower in nUlrients Ilnd highcr 111 fibn: Ihcn fishmeal. Combining soybean "'lIh cassa\"a increased lhe methionine defieier.cy. Thc advanlagc Ihal wybean had over fishmcaJ is Ihnl il ;s not easily eonlamlnalro wllh sand and waler. Also, as mcntioned before. fishmcal can be eonlaminattd ... ith Salmone\la. Hn"'avu diel5 in equal amounlS. There shou!d be no diffcrc n ~c belwecn yolk colours amongs l Ihe dilfcrcm Ircalmcnl ¡¡ruups. Another mClh04 10 increasc Ihe yo lk culour would he to includc. lhe more c~pcns,ve. )'clluw malze in lhe dlh nlles :md egg productiOll were lower rO( lhe eluan trnlmenlJ Ihan for lhe conlrol,. Thc l'Casons fur !h,S wcre prevlOUlIly mcnlloned. Howevcr, 1I became clear thal pcoplc COI.Ild be innucnccd hy the ouloome of dcmorulr.,ltlon lrialJ. 1lic owner uf Rolo (lIon, for cJCample. decidcd 10 ehllngc her fcal 10 thc: control rcal J ue lO lhe melU$C tn layer peñonnance_ For lhe demonSirJI,on Ingls 10 be 1$ effecTlve roc !he adopI;lon of c_v .. the dernonstrnuon ¡rills should be repe"ted wlth pelJeu:tcd cusa~'1 diets. prererably wpplcmcnlcd wi\h mcmlonllle. As mcnlloncd bdort:. pelleuud C:waVI dld¡ are upected 10 n::wlt in I highcT pooltry pc:ñormal"lte. ObiM' ... ·l tioll5 by Julián 8oitngo (eLA YUCA t"On'iultant ) on lhe t"e!IIu ll. obtallM.'d In the rfflUng trials • • • • The rcal"'g Inals w1lh hJgh lelel! of cassal-a rool nlelll conduclcd under T"nZllJlIan condu¡ons showed ermtle rcsuhs '" brotlers and laycrs, both in performance IS \/I'eU as", monaJlly. Ol'ernll peñorm:mce of broile~ unú laycrs was quite poor in all trials, ,ncludlng cornmcn::ial controls (final ",c,ghl al 6 "'ecks i8 slogJuly hllher Ih"n 1.0 kg, companod ",ilh 2.0 kg in commerciaJ broller produclion). MonuUly was qUltc hJgh In dletJ wnh hlgh Icvels of fi~hrnea1. In COOU"tlSI. higll lel'els (SO,*,) of Cllmov~ meaJ hnve demonstnlled e ... cdlent IlCñonnance ",hen mÜl,cd ",ilh OIher hlgh qoalily fccd ingredJent$ IR Olher Teg.¡OIlS . Aroller dlels ""ilh 50'1, cassava roo! me,,1 and 6% eass:wa lewf mclll re~Jtcd In cxceUenl "'elglll Illlin and fccd COnVelSJOn ","hen ml~cd wl lh (UUflll wybemns (whole ~) processed by roas1inll or eltltmlon. This I)"po: of wet5 dld nOl produce lhe llllfle resul~ in thc Tanzanian dc:monstnotions . Sume uf lhe hn1tllng fac lOlS may include. The Iype or fishmea l used in Tanl.llnta J~ ()f ponr nutritionul qualily (high IIIIh, hlgh falo 10111' pmtcin) and \/I,lh a high II\Jcrubilll con lBminalion. The open procening of fishmewl does not pro~ide the nUllImal sani laty condit ions and OpcM Ihe cnlruoce 10 SahnonelJa. E. Coli and CIOSlridlum. These m,cronrganisms ma)" originwle di!;(usel. whlch 111'111 arreel po:ñonnance and produce high mon"lity. 1ñc dlel5 wlth hlgh CAUI'''' lIfC ~ susceptible Iu m,c:robIll contammation SInce lhe)" WIU contatn hlgher IcveJs or fishmcaJ Tlle hl¡h ICl-el of fQt IR fishmeal 111'111 &Iso faclllt:lle O"'Ic.latiOO ami nmcidily o f\he dieto whlCh h"l' nc:g:aIl VC effect IR anomal fecdJ COllonseed cake (CSC) 15 not I rc('Ommcndcd Ingreet'ló 10 dcmOflSlrale mcthotb of inlensl"e produchon of cusava fohagc 3nd also I~ lcchnique 10 produce good qualily silage wuh ClISs.wa leavu. As a follow up of this acUYlly. l'OII1e tri3tS with milkin!; m .... ·s well' dcsigned. Thesc lrial~ w~~ In 8CI 8S a dcmonslralioo e rrecl. LOL promotC5 me prodoclion and ullllzation uf mitk in MuJawi Ihrough tcchnic;tl il$islance. ínlroduclion of Ilrtificlat !nsc:mina tion lechniquc5. e)(lension servlCCS an ll lnlining. LOL ,~ very keen lO H:chnology intcrvenlions and has h3d goOO rc5ulls in lhe pBS!. For cumple, usmg ímpro\'cd feed (moJ:lss.csj, mili: yictds h~YC ,ncrcus«1 fmm J !tI S liten. pcr day. FOlTlllllatiuns werc based on Ih~ u.~ of motass.cs. maize, cement and ot/ler raw materiab. Expcriences wilh cusa\"a were new fOl" them alld decided vay enlhusiurically 10 partidpate in lhe 1riI1s. "The Figure 2 presents sorne o fthe results obtaincd. Value. ' epie5erd me average daily milk prodUClion of) CO'NS IIIaI wtTt fed with eu5lva luves sil. and } CO'NS fed with maiulgrass silage. 80th groups were milked durina 80 COIIS«UI;yt dlyl. 1I CIII be obser.-ed IIIaI mi lk produetion of cows fed witb lhe cau..VI sil. WlS, ifnot !lUpcrlor al leUI eqw.I, IlIan lhe production oflhe cows fed with maiulgrast. This could be a very importan! oppommily 10 promote • more inlell5ive use of cusa\'&, roon !Il1d leaves, in lile dai ry sector of Malawi. These resu lu . a1though very encouraging. llave lo be taken wilh some caulion becluse lbey were obtained in Katete Farms. one of Ihe luges( milk producers in Malawi, uooer real life condilions, bul lhe scale oflhe trial was Imall. Wllat is imponan! is lhe indicalion tbese: re!IUlts are giving about me great potential of caS!la .... lO become , raw material ror lhe dairy sector. Funher lIudies, wilh more sdentific Oasis need 10 be conducl.ed in me fulure. " , ~ ~ ~ ~ ~ ~ ¡ ~ * ~ ~ ~ ~ ; ~ ~ ~ ~ ~ l MUkioK d.y. Fi¡ure 2. Performance of milkinl tOWI in Malawi (ed wilh CAssava ,ll'ge. b. Establishment o( al Juu olle pilo! projec:t .sin, tite inlqnlled produd developmtfll approub In 2002, lbe SARRNET leam in Tanzania iniliated a pilot project in lhe village Bunsu kxaled 140 km SOUlh ofDu es Salaam. This region was sclected as I potentiaJ area (or IllIll5ferring improved CUSlVl. prooessi"8 and nwketing lechnologies. bascd 00 J7 lhe difficulhU Ih:.t farmers "'ere foclllg to seU IOC cassava rools in ¡he D:II' es S:llaam marl.et. Cassava mKlers prefer !O huy cassava f'OOIS fmm fa1"11lCrs IOC!lled nean:r Ihe capital. mninly dllC to poor 11)00 cornmunicalinn hc,ween ¡he capital and [he far disl;¡m vi1l~ges IIke Bungu. Thls viUagc wu chosen a~ ¡OC pilO! projec[ arca considcring il. hlgh ca.sava produclion. high )ields (lO ~frlha). low peSI ;md disease pl'n'lure and IOC grow,ng importance of !he crup as a majar cash income for fanners sin<.:e cot'onUI aOO cashcw nUI produclion are decllnll1g in lhc region. Prn:es paid by caliSava Ir.ukrs in village~ near Dar es Salaam are as high ¡Q; tlSS 3OIton (March. 20(3), wllereas in regu.m. li ke !he Bungo village. ,he priccs pllld 10 c assa~a rarmen are aroond USS 1511011. farmers around Bungo villagc Il1Idiunnally C;ll IOC ca~va nxxs in fresh form or proces.sed in Ihe form uf 11 producl called Makopa (medium 10 large size chips dried fur 8 10 10 days). This long processing period produ~e~ SOfne fcrmcntalioo uf Ihe cassa.va chips and very oflen Ihe fioal product comai"s a 101 uf dust. fungi and soi l partic les. Onc uf the slrenglhs of SARRNET ",as precisely the aVallahili ly of chipping and drying tcchnology for cassa.va chips. 1llese technologle5 inc l ude~: a) a simplc chipper mOluri7-W " 'ilh a 3.5 hone power pelrol eogille. wllh capadl)" to produce 400 to 500 kg of fine ensava chips per hour, aOO with a 101:.1 COSI of around US$ 400 per unil: b) a drying mclhod usuaUy hascd 00 raised drylllg trays lhal allow reduclion of lile moi51ure in IOC chi ps to uJe Ie~els ",ithin Olle d:ly of e"J'05ure lO the sun. Thcsc lwu eomponents represeolcd lhc Icchnological inteo'eolion Ihat was (lIUCnlcd 10 IOC farmcr;s in Ihe Bungo vi!lage. TIte idea uf seni ,,! up ~ pilO! project ha.scd on this innovatioo .... as presemcd 10 lhe farmen during a publie dcmonslmllOll C(lnduCled Ul IIle villagc "'¡ Ih Ihc hdp uf local agricuhuraJ, cXlension sta« aOO runne ... leadcrs. During lhe demooslrdlion. SARRNET offieers broughl lo lile villugc lhe chipping unit and sorne dryiog Irays. Both men and women lesled Ihe equipmem and becarnot "CQ' interested in lile dcan and ... hite cassava chips produced. They ,,,ter took home the dricd chips and have lhem lesled In Ihei r Irtldi tion31 foods. Additiunally, SARRNET uffieers look lhe high quality CIISSIIV3 chips ;nto D;lt cs Salaam and cSlablished cuntacts wllh polenlial martcts thal became vcry inlereSled in the producto Afler Ihe $uccessful demonSlruúon. huilding upon the intereSI ~ho""'o by farmers, thc SARRNET tearo and coll:.boraltng inSlilulions in Taozania dcc idc:d 10 eslabh~h u smatl ' $Clle proccssing plant. as thc pi lO! proJCCI silc. ",ith Ihe Mpondi Fa~r and Business Group. a \'ery ",cll organized group of farmen cngllgcd in cassava producllon and very e3gcr lO gCI involved In cusava proccssiog In eomrnerdal sca le . Thc deal proposed 10 fam..: !'s was lhat SARRNET ",ould provldc Ihe drying Imys and lhe chipping una ",hereas thc farmers wiU be rcsponsible for butlding lhe prot'essing shed. The cighl mcmbel'1 of lile Mpondi groop conlnhulcd ",ith Illei. own cassav3 roo¡S and providcd lhe t.bour required for lhe processlIlg activilics. Vcry soon ¡hc pilol planl wa~ opc:ralmg and thc group was In huSlnel5 with urnan markels. J8 The fanners group n:'cei~ed direct, learmng-by-doing tr:llmng on me proce~~ong te<:hnology. Some of Ihe frumcf$ "'en: 1I1!oO lakcn 10 lile capilal 10 VISI I major markets and gel ocquainted with potcnual buyer.;. By Mil)' 2002, !he ¡;mur was rully operaung !he cas~ava procCSSlng pilut pl:mt with an (M,UPUI capacity of around 400 kg or high qu~hty ,assava chIps per day. A sigoifkant e.::onomic impoct ha~ beeo IICt"IIevcd wilh th lS simple. small-scaJe. affQrd¡¡ble Icchoologieal imcrvcnlion. Farmcrs an: now able 10 sell loor ~;c;~ava moI.S:1I tlener poX"eS 10 ¡he proo:rssing unit, " 'h,ch is loeated ncllr lhen cas.saV8 pIOIS. Bef<.IfC SA RRNtt Intervenliun, lhey lI"ere having d,fficulties selling rne cassava. TOO!.! since ¡he ooly avaolable tlllIl"tCI were fe ... trnders YI~,ting Ihe rt:gioo. The prices they reccived were ,.~ 1011". With !he ncw mmet. lhe prkes they rc:ce1Ve are higher and addlhomdly. Ihe proc<:551Og aclivilies at lhe pilol planl are crc~ting employmenl opponunities, especiall y for older people aod " 'ornt"n that lake care of peellng the rovI$ prior 10 chlpping aOO drying. Abo, OIher villagc:s .od 10<:31 Ir.tel al nationalle,·el. By Mllrth-April 200J, lhe SARRNET leam in Tnnzaniu. logclher Wilh collabor:uing 1081;tullOl15 wcre invite.:! 10 vi~it OIher regions. in Soulh Tanzan la , tO discuss wi!h regIonal and díAnCI leve! 1I1l1horilie. Ihe possihilitics of impkrnt"nting ~imlrar work, l"hc dlsc ussions nlO\·.:d forward 10 Ihe enenl Ihal plans are underway to establt~h new pilOl proJb;1ll in Ihl$ region. based on !he ~ame concept and tnClhodoloSY uscd 10 !he Bungo ~111age . 11 IS upe<;led IMI by !he ene.1 uf 200J, w' lh financia] support coming rrom 1m: local ilulhonties aOO dlSlncts. !hesc ncw pIlO! proJCCIS will be (uncliooing. A clear cumple of !he quick adoption Imu dirfusion of an IIll"lOvalion lechnology when 1\ its 39 meanlngful 10 lhe majn !:lern:ficiaries. lhe CIIS53\" f:umel1l. Anne:,; I prescnl5 addnional inform:u;on and sorne plCIUIt'S lhat iIIU.'lIrale lhe X UV'\les condUCled:1I the Bungo Village PiIOl Projcct. b.2. One 1'1101 projeet op..!Mlling in 1\11,111"'[ In Malawi. a ,imill!!' approoch "'a~ followcd to.se! up a pllOl rmjec! through which !he COlICept of lhe "lintmg farmc:rs to marl:eL~ Slr~legy" could be lested. Thc area sc lceted for lhe pllol project was Phalomt>c. one uf lhe 27 d, smCIS uf Malawi. localed 137 km e:w of Blanlyrc. lhe mUSl commen::ial city of lhe CUUnlry . In lhe regioll. lhe major SOIln:e of illCome for fal'Tl1ers is lhe sale of agnculturnl prodUCIS ~uch U5 m3i1.c. p.geon pea and sunnuwer. Tradel1l who establish lempocary pun'hasmg places dommote oommerclalizaliun of agncullural produclS. The agncultural prodLKIS are: transponed later tu the c1ly of Bluntyre for use in human consumplion and induS lri~1 markeli. Through a ~trategJ c alhance wilh lhe Christian Sen'lee Cummulet: (CSe). an intemaUOn31 NGO wilh opcnotions in MnlBwi. Ihe SARRNET leam In Malawi inlllnled lhe pilOl projcel in lhe Phalombe repon. Farmers from cighl Yllluges wcre Invited 10 pá!1icipale. The initlal conlacl wilh fBrtners was malle Ihrough a $Cnsilil.ation meeling held at Kolowito vil1age. allendcd by o~er 200 farmers. moslly women. Thc objecti\'e of this meeting was 10 sensilize fr.uners about lhe importallCe of cassllva as a eash crop. (Ol" food sccuri ly and also as n 51rategy for opening and slrenglhemng new markel5 for their agricultura] productS. 'Ole coneepl al using illlpro\ed processing teehnology 10 prodoced lmpruved quality prodOCIS wilh which 10 open new martet~ was dtscussed wuh Ihc farmers. SARRNET slaff brought over roor mOlonzed eólSSllVU chipper unl's and sorne dryi ng tnIys. Thls initiaJ meeling alSQ allowcd rarmen 10 pl:w::c a slrong ~mand rOl improved cnssa~a \'ariellcs lUId good qualily planting malerials. Later on. takmg advaJltagc: uf lhe e~islellCe of cassava phmling material nursenes In IWO research station~ (Chitedze ¡¡nd Kasonthula). SARRNET and CSC providcd plantong material of IWO recommc:ndcd ,·ariellcs. 100 furmers received enough planting malerial 10 plnnl aro nverage of 0.2 ha per famlly. In add\tioo. 15 demnnSlrdtion plOlS were planted using one impro~ed clone. The resu!t.s ohlruned with thc pilOl projccI appro¡lCh wen: nOl as enCOUTaglng as Ihose obtained in TanlUlnia. Allhough lite acceplunce of!he lechnology by the farrners was very good and the chips ublained were of impru\'ed qualil), compared with the t.radi ti onal ones. the faclor Ihit proved tu be the must hnming une W3! the very long dlstance helween Ihe pilOl projee¡ sitc alld lhe main mad lhat communicatcs ""Ih lhe uman móU'kc ts. TIle pri"ale sectOf"companies eontacted during lhe surve~s ,,"ere inleres¡ed in pun::hasmg Ihe ca$5a~a chips bu! they rcfllsed to eoUcet them al Lhe village. The hlgh CUSI of taking lhe chips from tIle ~ilIage 10 the main road and lalel lO the urban marteu mllde Ihe ""hule operalion "ery (llfficull. Although some few lon5 of cassa~a roots "'ere chlpped and sold, in general. lhe 1'1101 prOJCCI slrategy requ,red sorne IldjuSlmems. FOI nwnple. Ihe dncd chips can be slored 1M a place located on the mam mad, paying IU" 'cr lnon.portaliun COSI~: onee a sufficienl ~olume IS complel~d. a bU}'1:r could be sought Ih~t would he w¡Uing lu eoUeel ¡1J.e chips at Ihl5 place. Of COOf1iC. lhls w,lI mean ~dd¡lional bargaining abil ilies aOO powcr by ,hc farmers. 'JllC líflal price that fanne rs wi IJ recei"e for Ihe chips has lO accounl for tllC Irunsponallon COSIS bc:lwcen lhe pilol pro~1 sitc Hnd lhe main rnad sIurlIge place. As ¡»Irt of lhe pilol prnjecl work cOflducled in Malaw;, ~ome cass.ava produclion trials were ,mplemenled. Resul,s obtained are presented in Annc,; 2, e. ParlkiplIlion in tI~ proces.~ of lnmslation of the book "El uso df la yuClI Cfl la allmentadon anlnud", from Splluish lo t: flgllsh afld formatting ror ",eh publklltlon. This acll Ylty has beco COnduele:s and other syslems ¡hat can be ~un !lK'led tu obtain odd;\;onal informali!)n rclated [O ¡he POS! han-esl handling and processing oF cassa ... a. 42 vii, Bjbhogra(!hic refcre!!Ccs. This sccuon includes the reference~ LI~d in file dc vcloplJII:nt of Ihe s)'stem. The syslcm has been devcloped 10 enslIre a ... ide diffusion :tOO InfOlTlllluon IICCCSS 10 C3Ssa~a farmen lUId ~5~, local de~eloplJll:nf agents, fIIfal agroen!erpnses. NOO's allll go~emmt'nfal (k(:ision makers. The s)'s!em has been plllCed on Ihe websife of C/A T's Rural Agroenterpnse DeveloplJll:nf projecl fOf eonsuluulon using Ihe FronfPage software Mlh in Spamsll lUId English. The English \'er1I011 is a~ailabJe un o cd rom, TIle lronf page of!he: Iysfem is shown In Figure 3. J." Themt I V. Staitholdtrs lf'Qjn~d ¡" "t.w skill lo supporl th~ marI-ti dri~~1I slrollgy 3.4. ll ntrodLlclion PaT1icipution of CIAT and CI..A YUCA t~hnic:tl personne1 in fhe implemcnlauort of fhe present agrecmcnl "'as concel\'ed as an cxel'ClSI: In which. concepu. mfOlTlla! ion alld cxpcriencc~ gaincd throughout the lut :ro )'ear~ of work In L:ltin America. were 10 be sharcd wlm techmcal pcl'$Onncl froln l ITA and SARRJ\'ET. In the fin;¡ mstan.;e, and lo a luser eX!enl, wllh ltehn ical penonnel from lhe naliana! insll lutions and otller agencies 111 Tanunia and Malawi tha! are oollaborJling ""ilh SARRNET. J.4.2, Exp«too resull: Upduted knowledge arnongst kcy SARRNET stakeholden abool ca.~~va prWlKuon and processmg Icchnologies eXlsung m L:ll1n America aOO ..... ilh polenlilll tO be adnptw. 10 the specific coodilions of SQme Afri can COIInlrles. J.4.3, Acll,1tles implt'menléd : 1. Facts finding lIml planning tnJ'l 10 Afnea (Tanlanill, Malnwi), hy CIAT/C LA YUCA pc-rsonncJ (RlIpen Best. Bernardo Ospina ). November. 2000 2, Fi r>t consultancy missioll by CLA YUCA (Bernardo Ospina. Juh!n Bujlr.lgO).Februl\l'y, March. 2001 3. Trnlnlng COllnc in Agroenlerprisc Dc"cloprnenl, CIAT Instru<.:lof]¡ (Rupen Best. C:¡rlos Ostenog). May 2001 4 . Panicipation of CIAT (Rupcn Bu t) in Stcenng Cornmince Mcchng. TanUlni~ . May 2001 5. Secooo cunsuhancy mlUlOn by CLA YUCA (Bernardo Osplna. Juhán l:Iumago) . Oclot>er, No~embcr , 2001 6. SClenufic c~chnnge ml~~loo of SA RR l'.'ET pcnonnel lo l.:!lin Arncnca (Colombia and BrOl7jl). February-March. 2002 7. Participution of CIAT (Ru pcn Bes!) and C LAYUCA (BcrnanJo Ospina) m SIt!elÍng Commiuce Mccling. Preloria. APlll 2002 8. ThinJ consultancy mISSIOI' by CLA YUCA (Bernardo Ospina. JlIli~n BUltnlgo}. ApnJ·Ma)'. 2002 9 . Founh consull~nC~' ml SSl...., by CLA YUCA (BernanJo Ospina). March 20m. I.en...tuft()' ... _ ot nAT"~YUCAICC""" p'''OOMkIIlil 2 P .. "";p.t ... ÑOATa(.uyUCA pu_1 ,n SARR..'O!T s.,.11II C-_ M.)1 (rop " , " """,,,,,,.1 pcnooono:l o ... ~ .". ..... fA (N,l,lVI ano;! f"'Ht fccdj ... ,," SAARI'ffiT T~ CAT .... ' ."'- Rwpen IIeSIlnd Ba-.iu Or.p ... ¡wt.:1piUCd in !he s-nna Comnúnoc: Mccunp IN! lOOl pt.,. '" Tarw ... (10011 ..... I'tdor ... s...... A"'", (200'2) P" " .. ~. 1 .. be od.apIaIIA Afncln , on !he ' IUII'"IIXI """ '"""" 'aro enl.."., .. (II'OJ«U l>.,nrd 011 ptOjfCI delill'_ .~«u[),'" .00 lIIOIIilOrinl ... A:q!Of1 (If .. SIftj", .. """,",- ltqIoru 1. U ..... T "1' .q-o '" Rqoons 1. 11 and l. I "'''' ....... ........ . -...' .... ""T_~.",,,,,, __ p,.D ""_._' l" ...... tlo. Iy.t •• o- .. tn ......... " ...... "t .... 'roe."'''' o. c .... .. C.sw. •• s .... o, t ... mlKt npOOl ... t fuod _lnt ... ~S, ...... II .. _ ....... loun:. of C ...... S ro. ....... sao ........ ~. FOt tilos ... .on, di~_ .... ~ .,.¡ "" ..... 'oonaI .. t,t ... No .... o;IIdoc.otlOl '!9"'I'ic.,,1 ... ou-c •• _ .lfortJ 10 ckI~~.og and itJll)fOv.og th, poilh.rveu m.".m."r _ proc:""rtg of CH .... a ..... lh .... , 2C V .... Dotspil. the f.:t t".,.t ' Ktensr.-. Inform.lllM h.H b .... g"'-~!.cj on lhI!I ICIfIIC. mast of this InfOtn\atioto IJ ""t ...-y 'ICC •• 1lI* 11« ..... iI .. l 300m lhe t!xperienccs of ClA T on ea.\SRva developmcnt in Lalin Arnerica bUI alsu. ClA T and CLA YUCA technical personnc! hayc gained ;nvaluable expcricnce abo ul Ihe currenl sltu.Uion ofcassav3 in !hl ! part or Africa. Ihe challengcs and ¡he o[lf1'Onumlies. In general il can be said Ihal sorne of Ihe key sllIkehulders uf SARRNET are now bel1er infOl'Tr'lCci aboul lhe pOfl::nllal opponunities lhal can be dcrived "y arrl)'lng a markCI drhcn approach 10 cnssava research and dcvelopmc nl acl;Y;lie!, 4.0. Cone lusions nnd recomnK'ndnllo ns 4.1 Inlrooudlon CIAT 1IIld CLA YUCA were ¡n~iled by IITA-SARRNET 10 provhk conl'cplUaI and pr.IClical inpul 11'110 lhe process of !"Cahgning SARRNET from 11 predominanlly producllon focus aimed al meeling footl securi ly obJCctives 10 a more m:ui:el 300 enlerprise orientaríon. \>I,lh a view lO SIIlisrymg ¡he illCTC:tSmg need rOl' providjng in.comc: and cmplo)'menl generating opponunllle.~ r,.,.. CISSIIva and swc:ct poIalo rarmel'S. 4.2 I\hjor IIchievl'llIl'nl.'i The major achicvcmenls ovcr the pcnod Novcm hocr 2000 to Sepiember 2003 ha~c tJec,n lhe rollowlOg: • Thtmt! IV. Stobhllldt!1'"f lraóntd ¡n " t lo' ,kills lo supporr ,he mQrkn dril."" stratt:KY. Adhil' ....... ' .. I.C"""'I .. ....,. ...... oons o( nA T .teLA YUCA Lechn",~1 ",.,OMe! 110 2 PlftlClpaIJOCI orCIAT.tCU. YllCA ""' ..... ocI iIlSARRN~"'T Stann, Comm,n ... Mcet.np ; ' ptOp ",ilh R"""" B"I'I and aerlW'do (}opina patI;"ipa¡ed in !he Su.ell"¡ ConllJlil1oc Moeung.< th.o.'loolr. plil<'C in T arw.m. (2()()1) • ...;! !>re,,,, .... . Soulh Afri~. (2002) , I I i j, ""tI> SAR RNT " . ir>ed on ,he Iaooc Repon uf [he Stceri"l C,ummllla: ' ¡h:p""-' l . rr . nd 1, n and . 1Id """,,,,,,¡ni (lf "Xli ogro Final , ... " inf'''''''''M''' nn 1otIa~114 ,'-' pn.>rmIe lIII,ket-· of lhe Iledgling prodllClion and proces$ing entcrpriscs and rheir respeclive supply c hains. R&D o n marl;cllng und proccssing has 10 go hand·,n. hand wilh complementllr)' and wcll -targeted agronomic research and the pmvi~,on of dcvc lopmcnl Strviccs such a.'! secd multipliclltion and other input supply. 4. An cnlrepreTII:urial spinl 10 progr.ml dcvelopmenl IS cS:lCnllld. Thls mean5 bemg pro8Cuvc in approoching Ihc privalC seclor, nOl onl), in ICrtTl$ of potcnlia l mar!o;cls. but also in ICrtTlS of cngagmg Ip¡n-opnalc Icchnologl<:al 5Olullons. Immen5e benefit ean be obt:Uncd from Soulh-Soulh cxchangc of mform:lllon and lochnology. wllh lile po5$lbi lllY of cnhancmg lr.wc aJTlOIlg conlim:nls. Both publ ie ;m¡J privalc seclor pel'5OOnel should be mOl,valcd (and pm18ps facllllalcd) 10 10010: fOf novel approoches beyond nal lonal and regional hound:lries. S. The presenl proJCCI provickd a focused and I1lC.1ningful opponumly tO m31o:c: use: of lhe eomp1cmemary ~ k, lI! and comparJ1ive ad"anúlgc or IWO iOlemutiooal agncultul1Il rl:sclln:h ~cntc~. Much has bccn lcamed aboul Ihe appropriule mechonisms r(H' II(:hlev¡ng Ihls Iype of cooperatioo. Among (he maSl impotluot le5sons are: a. Despile the relatlonshi p belllg contractual io nature. more ;s lII.'I.'omplished Ir syocrgy nmong lhe panicipaling insüllIlions CHn foslcrcd by cngcndering a c ll m~le of mutual respc:c l and tros!. This creates aclimate of joint Ol>mership of Ihe process. b. Whcn Ih is rl:spc:Cl nr lru ~t is qucSlioned, potential impact can be consio.lerably curtailet.l , ÍIS the mOlivation for goi ng beyond Ihe immedialc lemlS of a conmlCl are reduced; ~. RcsponsibillllCS and recogni tion of SIICCessc:S 3nd fai lurr:s muSl be shan:doped. HiP u .. ~ 03lIl mdc " \'O')' dilf1CUl11O IIdI "....... molo n <;UM\~ produc:lS Ut ..ro...~. UsualIY. bnuo:n ~ 10 telJ Ihcir IwvaIIlI miocIdIesrICII ). farmctl lISClnIIiaIy u.2itional ,-aricties al1b ''''"1 ¡illle e.1ema1 ....,.n fcnili_ \lOe m vIn;oally IIOO~ )o T~ _ .. pOCU6liQ&: 1CC1uIoIo&ics are gommon me! pnxIocu otaaollCd (ej Makgpo .." ...... dlillJ) U\'C poor qtW~y lInd In: lOId ti ~ Iow pncos )o Fanno:n """' _l)' indlvidoalJy ",111 ''CfY hUle b.piniQ&: poooe-. '0 • 'fII.r Jl lervel lto. ~Ttc:bpoJOIj(8l.pd orJI-pu..,tioal! ¡aaoulio.'" ,. lIT 11. ... • .. !.-dI_ ."' ... .., ...... II:IMtiQ: .. Africa.1IIIII:d .. OOp¡owtd •• ...... 'P·for....a ........... .... ............ iI:I TkK ~ MIl IIDI bc.1.-d" T ....... bcfcft. ). CIAT lIId CLII. YUCII. MIl ............. oI.IIt...... ... jo¡¡ in l..PIlI.lacrica oItlrc -U .... F ar--. 10 MMIu:fJ" ancIlbc "11IlCpIed C ... Oe-we ............. _ • .......... - b. P;1of,... (l'Idd N/u/ '~and~"""'" i_J ). IITII. ancI SAARNET, .. i lh _'I.IAC)" IUppon from CI .... T-' CLII. Vl.JCII... tb1ed. pilox pn:ljoa .. "it~ l'áItw poapI in BIIIp. R1dI.ii ditIricI..1IIIp' _"1"""" rq;mi.T.,.... )o Ke) llIkchoIIkn Ir.- P"OUPf. NAlI.S) puuapMlld lIIlMIy ID W pilox ~ ~"IhomaI_ p...-n- P'O'"óded bJ lIT lI.ISAJUtN"ET _ ..... _.- ). S. ',*' 1-' ,i4dtbt ---........ --- " A. T~_,_ Rnull . o( lb .. Pilol Pbm Tb .. Ad.pU-llon IIlgh quali!y (hipos IIR IO"lIne. dall and an k lIICd KI rcpIac:c p¡utlllly ,,·haI! and mau.e lIIed ;". tbo ¡:.cpoo Ilion '" tucu.its. hoIIIe bIkcd pmduo::is Md upll Dl)ll11 "'"'"' .. minicbqJI 011 ....,,;1 dr)illl racb.\"Oids r~lOI\.-.d long peOO:III 01 tl'po5l1te 10 IIOi I Ill>Cl dtw contlmillll!lOII. final q ... lny of lhe prodPC1 is '~. 8ood C-.. ¡:.00CiSi1l& aan-.iel; Wt r.iliIaod Wllh \be .. of tk prOIOI)1!C" ~ 1!$Cad of chlppin¡ the_ witll kni~ fanners..,.. __ cfti¡)per run "itb _ pc!roI cngine. Their dJ"1Qe1l(y Iw iInpro\-~ IfaIIty The <¡UIIi1)' of Ihc dnl'" (gc:omeuy. Ii>Jl! unifonni1)'. tIKkncsi) iII lIcnet .,.¡ dl)illl !Ulle has beaI mIui;al IMead al IIdIIn¡ !he t:aIA\"I roou individuaIty 10 mlddlemm. fanrICtos In no..- abIc ID ........ '*"CiaIilo: lIICir =P uanslonned n.o IIIp ..,.u.y ~¡:.. in Iorp YOIIUIICI aIId 11 .. ¡pe prioIs BcoideIIhe \:Cndiu; d in <:QMofIld!o;Iion ltChiiOlocY ....,.....,. .. üI .. 12 COI! offrall eaMII" (J J qfiTsJop I tr "'po. 10 TU. q""'ÑIJ --(3 pc:IJIl 1 'da)' • 1000T.wpa... 1,000 q ck)' drM .. 2 dI)'l) """'-¡n,"i('W,on,jNt .!tap. "'-. ~It) Rn.to: e(tltc f'üot ... _ En_K batdib of dlt i._.úo. {TsaIq dry ( hipo' Coa al processing (1 kg of dTy rossa>;Q cNfJf) Cost Or uansporwion (J5Tsil / qdry ~d"¡n) Currcat tdlins price (Delivenod iII Dar es ...... , (fal ka dsy cassa> .. chips) (fllh I ka dry ca$$8"" chips) ¡Tdl l ka) Thc bcncfits for WmeI'1 patlic:ipaling in the dI)' _ ... chips operatioa accrIIC ñum; a. Highcr pri<:e ror roob: b. EmpIO)"lnme opponullllics for cldcr poopIc and "'On>eIl ~ . la. dcpcndcncc OIIlradcrs do CaMa'" oIdcr!han 1 ycar hu. nwh1. e. Oood quaJity dry cau;l.va dupl" aJJoy,ing thc:m IOmler urban maM > Frcsb Mmct """- Ykldofl~l / ha) FI'C$h marlr.ct pricc T$I\ ISt lq¡ 01)' easslVllo d!ips """'. (roou IOId al \he: prooc:J$I"i plan! gau:) Assumin¡' ). Yield ofl S t l ha ,. Prooeu'''i planl pnc.: 20 Tsh I ka (Owncn of lhe PIO<.:CA'n.¡ plant) ,. PlOCCISlRg costs of m..s T sh I k& dry &a$$II\1I chips 'lO rr"""', 15,000 ~ 13 (Tsh l ila) 15,000 x 20 DIle group of CUsaVI farmers in Taruani. can now, willl Ihe technologiCIIJ ,00 organizational innovllion obtain«l developed by SARRNE'T, estabJish l different canavI production, procusing 100 commerci.Jisatioo 'yAern. For example, ¡flO farmers gel. together and organiz.e. dry C&Ul!YI chips CUS/lva-processing plant, and caeh flllnCf scHs 10 lile proce$Sing planl 15 t of cassava rootJ., tbe planl will be ablt 10 opentt 100 day' a )'Uf, will proteS$ 100 I of dry 1:4151.'" chips (1 I per day) and net profit wouId be around 4.95Og d!stancc¡,: (Ix 1); (O.S A 0.5); (0.9 ~ U.50) and (O.JO A O.JO) melen Th= repetitions. pl:mling date: Oclober lOO. Four har,c.st periods: 15.24. 44 and 56 wccu after planhng Arca harve.o; led: 15.6 sq.ml li ur"cSI sySlem: ratooning Table 5. Cas.~ava leave!i Iríais in Mllluwl- 200 1- 2002. Katoonln¡.: har>"e!it syslem . " • " " " " " 1" " " " , • " , .SlI .. • .5 ••. s o." • .5 (U1'.J f""'l"n' 4. e ....... lu...,. p.odllcCioll trial ¡. Mal.wi. Ratoolli"l syslnll. lotal ~Id ill fin b.~ . hl •• .5 •• .5 e'., . 0.5 DO.J.U .. .... ..> " "" "'1 .1 ". ,» " " " • .. 1 " " " .. , " 5111 .... __ "i M.njl" V.,wlia f''1""'5.. ea.u ... k.Ya prodllcCin. lrial .. MII.wi. Ralooailll J)'su-. total ~Id ¡. fi.,. barYftU. ss CllSIlIIva 1fi¡,.1':i pruductklll Ir i.aI- llt loPlling sy51em The experimental wor~ Wl lh mleOSIIava Jcaves prodoclion systems. especially In c~s whcn: suppkmentary irrigation 1$ 001 nvniJable. • From lhe data obtaiocd il cun be obscfved !hat eas~va plants grow sutisfaelorily up 10 Ihe first harvest. then , wnh the dry season seuiog in. plant growth is .Imosl Slopped aod $Cvcral monlh.~ la[er. wah more r.linfall. plaol growlh is recupcrntcd. l'his sugge~1 thal In vel)' dry !'egJons. wuh more than 5-6 mooths of very 101'.' I':\l orall. the establishment of farmmg systcms for ;olensive prodoction of caS5aV~ 1e3Vt.S could be "ery rhky. ea. ... ·¡¡ .. ·1I roots productivo Ir iMI TlIk ing advanlage of the cxisti ng experimcnllll plols for C8ssava leaves pmducuon systcnu; tlw:: SARRNET team in Malaw, eondueled anOlher CXpcrimcnl with lhe ubjecti,'e of looklng al rOO! yleld of cassava plants lI.rtcr the last han.·cst or the fo ll age, and companng it " 'nh s,milar plOlS ¡ha¡ huve bec n lepl un harvestcd throoghoul lhe pcnod of the expc:rimcn¡ (72 weeb. 18 mootM). Table 7 and Figure 8 prescnt lhe dJIJ obtainetl " T.bIe: 7. Yieleb o( ( .... VI produC1ioa I)"Slt •• ln hrO ,rowi", (yda. /\ohl ...... 2001 · 2003 . • Maunji,; (I~I) . Maunji'; (0.5 .0.5) .. D Maunji,; (0.9 .0.5) D Maunji,; (003 I 0.3) "---~----~----~~--~~, . .. "" " • Fi&II~ S. Yldd. (Ir tUSava prodllclion 11'11'1111 in hrO ,rv .. i", C"ydl'S. Mala .. ," 2001 . 200). 62 'The results obt.ined 11ft' vcry tnteresllng. 1 t can be noticc:d lhat despi te lhe f OCI tha! p)¡¡nUi were han'Qiled live lilllCll. during lhe 1:lSI h3J'\'est CUfHJUCled 67 nlOfllh.\ 131er. Ihcre was sli11 a Slgmlicallve prodoclion of l'OOI$. Dala on dry maner of lhesc: ruoUi was no! available al lhe momenl of ~panng th ls o:port, !IO !he inform:llion of lhe qualuy of Ihe rools.1 this momenl is rnJssmg. Howcvcr. >,clds obtained wen: gQO(! and Ihls ~uggesls a crucial queSlion: is il bclter tu lellhe c;u¡~a"a crup grow for lWO eyc les mSlead uf une? In a counlry like Malawi, wilh a shnrt rmnfall seasun Ilnd 101\8 dry s.eawn, whal lhis prelimlnary cXpc'rience is ~uggesling is lhal cassava planlS when allo"'ed lo ha .. c lWO rainy periodJ. nver a growing cycle o f 1 S lO 18 monlhs, cnuld give Ycry good yiclds (4{)- 50 1 I ha). Thi5 I)'P" of erop manDgemcnl i! procllced widely '" advancc:d ca<..~:wa prodochon syslems such lIS Soulh of Brazll , wilh " 'cr,lgc )'lcld5 of 30 -W 11 ha. To corrobor .. le Ihis, :motocr pie<:e of m(ormalion was obtwnc¡J rec.:cnlly by SARRNET- Malaw¡' h sc:cms 10 indlcale lhe same type uf data. In Novembcr 2001. three farmers in Malawi dCClded lo inves l in a commcrcial scale casSllYa-grnwlng schemc. und plamcd SS has al a flllTT110calcd m lhe oulskins of 1~longwe (Chitipi Fllrms). Whcn it was lime tn harvesl (Seplembcr 2(02), lhere was a huge dcmand for canava planting matenal in Mala"" and milny in!emationaJly-fllnded projccts. NOOs ano:! even local in5tiluliun~ ended up pun:hasing cassava cutungs (mm these Ihree fanneB. They ma.de a 101 01 mooey I'IlOOlllng the cmp Iha! wa~ presenling 3 vigorotl5 Ue"elopmenL The crop w~s lel 10 grow ror another cydc. Reeenlly. SARR]\'ET dld an CllerclSC: 111 thlS rarm 10 dclerminc quality and yieh.b of lhe cmp. Wilh plunting dislallCes or 0 .90 ~ 0.70 ( 15.873 pl~nl~ I ha). the ylcld ils~ssmenl done hy SRRNET-Malawi indicalcd an e~pecled yicld of 39. I I per ha. The dala oblained with ~11!jSUV~ planUi growing for lwO c}'cles deseNe~ 10 be inveSlig:lleu rurther. 1t couh1 be un "!ternalive lo deve lnp ond eRtahlish more compclilh'c. cfficlent und 5uslainable cusa"a furmmg systems. 63 • ANNI:X 1 CASSAVA PRODUCTION IN TANZANIA · FlELD T RlALS SARRNET personnel in TanDni. al50 cstablished. in elose coI labol"llion wilh technical penonnel from local institutions., • seI of lriab 10 1$ varic!;ies and crup management prKIiccs fo.. both. sweet powo and CUSlV' (rooll and IeIoves). 1ñe aim of these lri.ls wu 10 determine Úle potentiaJ yield gaps for both crops. when production packages IhaI include extemal componenu are used These componen15 are mainly lhe fatilizers and tbe improved varieties. The comparison was made Ilsing lhe local, tJw:Iitional aopping syllen! as lhe check Data otuined in lhese lria!S and lOme of lhe preliminuy conclusonl are prexnted as follows· TreatmenlS. One ~y (Eliasi) Four fl'ftilization methods: micken mawre. canje manure., NPK and whhool fl'ftiliUliOfl Planlingdate; May 14· , 2001. Har"elldale: <>aober2"". 200 . Three repeUlions " ~-,... " J • , • • • • ,¡,. , " • • • ~ ~- Figure 9. Sweet potato ft'rtilil..ltion t rial., Tan;r.a nia, 2001. 64 SW~ polt lO fertiliulion Iri ... 2oo2 Treallnerus, thrH varletie$: Eliasi, SimlBa, Uker_e Four reniliution methods. dUden lTWIllI'e, ~nle minore, "'PK and wilhoul renilization Planting date' April 4", 2002 Harvesl date: August, 30th, 2002 Ares. harveRed: 42 !I