The African e-Journals Project has digitized full text of articles of eleven social science and humanities journals.   This item is from the digital archive maintained by Michigan State University Library. Find more at: http://digital.lib.msu.edu/projects/africanjournals/

Available through a partnership with

Scroll down to read the article.

Zomte/a (1988), XV (ii).

RESEARCH REPORT

A WEED SURVEY OF ARABLE LANDS OF THESMALL-SCALE FARMING SECTOR OF ZIMBABWE

O. A. CHIVINGE

Department of Crop Science, University of Zimbabwe

There has been no national weed survey of weeds of arable lands in thesmall-scale non-commercial (SSNC) farming sector of Zimbabwe. All previoussurveys done in the country covered the large-scale commercial (LSC) farmingsector (Rattray and Wild, 1955; Soane and Waister, 1963; Thomas, 1970; Budd,1975; Chivinge, 1983). However, periodic weed surveys are necessary as weedpopulations, compositions and intensity of infestations change with climaticvariations and agricultural practices (Chancellor and Froud-Williams, 1984) Asthere has been no previous weed survey it was decided to conduct a national weedsurvey of the SSNC farming sector with the following objectives:

(i) to find out which weeds are the most aggressive and difficult tocontrol in the different administrative and ecological regions;

(ii) to determine if there are any differences in the weed spectra in the fiveecological regions;

(iii) to find out which weeds need urgent detailed studies; and(iv) to establish a base for future weed surveys at national, provincial,

district and regional level so as to monitor weed pressure orpopulation changes every few years.

MATERIALS AND METHODS

The survey was done in two stages. During the first stage a total of 600questionnaires were sent to the Department of Agricultural, Technical andExtension Services (Agritex) and agro-chemical personnel from September 1983.The same exercise was repeated during 1983 to 1984. This period was chosen soas to cover both the summer and winter crop-growing seasons. The technicalofficers were asked to report on weeds specifically found in the SSNC farmingareas of that part of the province in which they were working. Those completingthe questionnaires were asked to list the weed species on the basis of aggressivenessof weeds and the difficulty in controlling them, noting any new weeds that hadbecome problematic in the past five years.

Aggressiveness refers to fast and vigorous growth in the early stages of theweed so as to adversely affect the growth and development of the crop in its youngstages. Difficult to control refers to situations in which weed control by handpulling, hoeing or use of ox-drawn cultivators is not easily accomplished or toweeds which quickly produced subsequent generations soon after the removal ofthe same weed by the above-mentioned methods.

The second stage was done from 1983 to 1985 covering both the summer and

167

168 WEED SURVEY

winter crop-growing seasons. This stage of the survey was considered necessary toverify the information from the first stage of the survey. The period 1983 to 1985covered relatively dry and wet years. During this period the author went to twoareas in each province and recorded weeds found in the fields.

For the selection of farmers whose fields were sampled, stratified randomsampling was used. Agritex officers were asked to divide each province into eightequal parts by area. Each area was further sub-divided into sixteen smaller areas.Of the sixteen smaller portions each was sub-divided again into sixty-fourvillages. From each village the names of one hundred above-average and onehundred below-average farmers were submitted. The eight and sixteen smallerareas were numbered. By use of random number tables the first number to appearwas chosen. The same process was repeated for the sixty-four villages. Then tenfarmers were randomly chosen from the one hundred above-average and anotherten chosen from the below-average farmers. Their fields of maize, cotton,sorghum and pearl millet were sampled.

For the actual sampling spot of the field systematic sampling was used. Theland was divided diagonally and sampling was done every 20 metres followingthe diagonal lines. A total of six quadrats were taken in each field and the quadratswere lm x lm. Weeds in each quadrat were identified and classified as'aggressive' or 'difficult to control'.

Each area was visited eight times in the period 1983/4 1984/5. The first visitwas at the beginning of each growing season and the second at the end of eachseason. The author also had personal interviews with some of the farmers whosefields were visited.

RESULTS

The administrative provinces and ecological regions are shown in Figure 1 Frominformation from the survey the weeds were grouped into 'aggressive' and'difficult to control' weeds on a provincial basis. Weeds were listed in order offrequency of occurrence.

The most aggressive weeds

Acanthospermum hispidum was the most aggressive weed in six out of the eightprovinces (Table I). The same weed was second in Mashonaland West and thirdin Mashonaland East. During the field survey this weed was found to infest mostfields irrespective of crop or soil type, although it was more associated with sandy,poor soils. In both Mashonaland East and Mashonaland West Eleusine indicawas the most aggressive weed. In lands which had been fertilized with either cattlemanure or inorganic fertilizers the weed grew very fast to the extent that mostcereal crops were easily choked, though all types suffered when infestations wereheavy. In Manicaland and Masvingo E. indica was the second most aggressiveweed.

Other weeds which were particularly aggressive in all the provinces wereCommelina benghalensis, Striga asiatica, Striga spp.. Tagetes minuta, Bidenspilosa, Rottboellia cochinchinensis, Amaranthus hybridus, Cynodon dactylon

O. A. CHIVINGE 169

Figure 1: PROVINCIAL BOUNDARIES, ECOLOGICAL REGIONS ANDFARMING AREAS OF ZIMBABWE

PROVINCIAL BOUNDARY

I

Ha

lib

III

IV

V

Related Farming System

1 700 m above sea level. Rainfall above 1 000 mm p.a. withprecipitation every month. Diversified farming

Rainfall 750-1 000 mm. Intensive crop and livestock farming

Rainfall 750-1 000 mm. Intensive crop and animal production.Occurrence of dry spells or shorter rainy season

Rainfall 650-800 mm. Semi-intensive farming with mid-season drought

Rainfall 450-800 mm. Penodic droughts and severe mid-season droughts

Rainfall below 560 mm. Very erratic even for drought resistant crops butgood for livestock.

Table 1THE MOST AGGRESSIVE WEEDS IN EACH PROVINCE*

Mankaland Mashonaland Central Mashonaland East Mashonaland West

Acanthospermum hispidum 18Eleusine indica 11Bidens pilosa 10Commelina benghalensis 8Cynodon dactylon 8Amaranthus hybridus 5Striga asiatica 5Striga spp. 5Tagetes minuta 4Eragrostis aspera 3Nicandra physalodes 3Cleome monophylla 3Richardia scabra 3Ceratotheca sesamoides 3Cyperus esculenlus 2Cyperus rotundus 2Hibisus meeusei 2Vemonia cinerea 2Vernonia poskeana 2Urochloa panicoides 2Ocimum canum 2Euphorbia heterophylla 2Gisekia africana 2Crotolaria laburnifolia 2

Acanthospermum hispidumSlriga asiaticaStriga spp.Amaranthus hybridusTagetes minutaEleusine indicaCommelina benghalensisRhynchelytrum repensCyperus esculentusPanicum maximumNicandra physalodesRoltboellia conchinchinensisBidens pilosaCyperus rotundusEragrostis asperaHibiscus meeusiGalinsoga parvifloraCelosia trigynaConyza floribundaDonyza sumatrensisXanthium slrumariumDigitaria spp.Ocimum canumSidaalba

Trichodesma zeylanicumGisekia africanaAergemone mexicana

29 Eleusine indica 2527 Commelina bengalensis 2021 Acanlhospermum hispidum 1818 Richardia scabra 1715 Striga asiatica 1513 Slriga spp. 1513 Rottboellia cochinchinensis 612 Tagetes minuta 69 Amaranthus hybridus 49 Nicandra physalodes 48 Cyperus esculentus 38 Bidens pilosa 24 Boerhavia diffusa 24 Galinsoga parvijlora 23 Rhynchelytrum repens 23 Sida cordifolia 23 Urochloa panicoides 22 Vernonia cinerea 22 Vernonia poskeana 22 Hibiscus trionum 22 Hibiscus meeusei2 Ocimum canum2 Digitaria spp.2 Trichodesma zeylanicum1 Crotolaria laburnifolia1 Gisekia africana1 Aergemone mexicana

Oxygonum sinuatum

Eleusine indica 24Acanthospermum hispidum 22Commelina benghalensis 18Richardia scabra 15Ipomoea plebia 12Cyperus esculentus 11Tagetes minuta 6Amaranthus hybridus 5Rotlboellia cochinchinensis 5Striga asiatica 5Striga spp. 5Cyperus rotundus 4Nicandra physalodes 4Portulaca oteracea 4Physallis angulata 4Urochloa panicoides 3Eragrostis aspera 3Bidens pilosa 3Celosia trigyna 2Euphorbia heterophylla 2Ocimum canum 2Sida alba 2Oxygonum sinuatum 1

mmD

Midlands Malabeleland North and South Masvingo

Acanthospermum hispidum 29Striga asialica 24Striga spp. 22Eleusine indica 21Commelinu benghalensis 16Tagetes minula 11Amaranthus hybridus 8Cyperus esculentus 8Bidens pilosa 5Nicandra physalodes 4Cyperus rotundus 4Galinsoga parviflora 2Portulaca oleracea 2Rhyncheltrum repens 2Rottboellia conchinchinensis 1Ipomoea plebiaAergemone mexicanaHibiscus meeuseiGisekia africanaOxygonum sinuatumEragrostis aspera

Acanthospermum hispidumAmaranthus hybridusBidens pilosaCommelina benghalensisCyperus esculentusTagetes minulaEleusine indicaPortulaca oleraceaStriga asiaticaStriga spp.Nicandra physalodesRottboellia cochinchinensisCyperus rotundusGalinsoga parvifloraOcimum canumJacquemontia tamnifolia

21 Acanthospermum hispidum 2518 Eleusine indica 2215 Striga asiatica 2013 Striga spp. 1812 Commelina benghalensis 1712 Ipomoea plebia 1011 Ipomoea spp. 811 Portulaca oleracea 86 Bidens pilosa 76 Tagetes minula 75 Cyperus rotundus 55 Boerhavia diffusa 54 Tribulus lerrestris 43 Amaranthus hybridus 43 Cleome monophylla 41 Celosia trigyna 4

Hibiscus meeusei 4Gisekia africana 3Oxalis latifolia 3Ceratotheca sesmoides 2Commelina forskaolaei 1Eragrostis aspera 1

OI

om

* The most aggressive weeds are at the top of the list and the least aggressive at the bottom of each list. The numbers indicate the number of timesthe weed was reported and observed to be an aggressive weed.

172 WEED SURVEY

Cyperus rotundus, Cyperus esculentus, Richardia scabra and Ipomoea plebeia.However, the aggressiveness of each weed species varied slightly from oneprovince to another. B. pilosa was the third most aggressive weed in Manicaland,Matabeleland North and Matabeleland South. This weed was found growingluxuriantly wherever there were heavy infestations. Well-fertilized fields seemedto encourage the weed to grow well, even under heavy shading from crops such asmaize and cotton. R. scabra was the fourth most aggressive weed in MashonalandEast and Mashonaland West. Information gathered during the field surveyrevealed that this weed has been on the increase in all the provinces since aboutten years ago.

C. esculentus and / plebeia were particularly aggressive in Mashonaland West.Farmers interviewed revealed that the main method of controlling C. esculentuswas hoeing which only removed the shoot portion. This leaves the tuberunderground which would produce another shoot. In fact, removal of one shootstimulates dormant buds in the tuber, resulting in more shoots sprouting. /. plebeiaand other Ipomoea species, were a big problem in Masvingo where they werereported and observed to climb on almost every other plant, crop or weedgrowing close to them. Ipomoea spp. were not easily distinguished from manycrops, particularly broadleaf crops in the early stages, because they were found tobe growing under heavy shading from crops. During that stage they formed long,thin vines which entangled the crop and would only emerge above the cropcanopy towards flowering.

In Manicaland C. dactylon was the fifth most aggressive weed. However, inall other provinces its frequency was quite low. Most farmers interviewed in otherprovinces reported that once the land is ploughed and the weed is exposed to thedry weather it easily succumbs to the hot temperatures. Manicaland is a ratherwet province compared to all the others. Another weed which was ratherrestricted to one province was R. cochinchinensis. It was reported to be amongthe most aggressive weeds in Mashonaland Central but was less aggressive inother provinces and was not even listed in Masvingo and Manicaland. In all areaswhere it was observed during the field survey it was found only in heavy soils andmostly associated with the maize crop. The growth habit and appearance of thesetwo plants are so similar as to make distinction very difficult, particularly duringthe first few weeks of growth.

S. asiatica and other Striga species were reported and found to infest maize,sorghum and pearl millet in all the eight provinces. This weed was the second mostaggressive weed in Mashonaland Central and the third most aggressive weed in theMidlands and Masvingo. Field observations revealed severe infestations anddamage to late planted and poorly fertilized crops. Many plants wilted underadequate moisture even before the Striga emerged above ground. C. benghalensiswas the second most aggressive weed in Mashonaland East and the third inMashonaland West. In the other six remaining provinces, it had more or less thesame frequency. This weed was found growing in all soil types and altitudes andseemed to tolerate shading from most crops.

Of the most aggressive weeds A. hybridus was the second most aggressiveweed in both provinces of Matabeleland and the fourth most aggressive weed inMashonaland Central. Least aggressive weeds varied considerably from oneprovince to another with only Oxygonum sinuatum being the least aggressive in

O. A. CHIVINGE 173

more than one province. Consequently, these weeds will not be discussed anyfurther.

The most difficult weeds to control

The most difficult weed to control in all the provinces, except in MashonalandCentral, was A. hispidum (Table II). Most farmers who were interviewedreported that the weed grew in dense stands and subsequent generations alwayscame up soon after removal of one generation. In Mashonaland CentralC. benghalensis was the most difficult weed to control, and it was the thirdmost difficult weed to control in Mashonaland East. Other weeds with highfrequencies included E. indica, S. asiatica, Striga spp., B. pilosa, C. esculentus,C. rotundus, A. hybridus, R. cochinchinensis and T. minuta. E. indica was thesecond most difficult weed to control in Manicaland, Mashonaland East andMasvingo.

S. asiatica and Striga spp. were considered difficult weeds to control in all theother provinces except in Mashonaland West. 5. asiatica was the third mostdifficult weed to control in Manicaland and Masvingo. Large patches were foundparticularly in Masvingo with the heaviest infestations being on pearl millet. Theonly province where R. cochinchinensis was considered a difficult weed tocontrol was Mashonaland East. During the field survey it was observed that thisweed was restricted to heavy red and black soils only. In cases when the weed wasleft in the field and had attained a height of over 30 cm, it became very difficult toremove it by hand pulling, hoeing or use of ox-drawn cultivators, as the weed haddeveloped strong brace roots. This weed was not even mentioned by therespondents in Manicaland and Masvingo provinces.

C. rotundus and C. esculentus were the only two sedges reported andobserved to be difficult weeds to control; C. esculentus was the third most difficultweed to control in Mashonaland West and the fourth in Mashonaland East. Thisweed was mostly found in lighter soils. T. minuta was more of a problem in theMidlands while A. hybridus was difficult to control in Masvingo. The leastdifficult weeds to control were different in each province with the exception ofHibiscus meeusei which was at the bottom of the list in both Manicaland andMasvingo.

DISCUSSION

The survey revealed that R. scabra, Richardia brasiliensis and Trichodesmazeylanicum were becoming problem weeds countrywide. T. zeylanicum germi-nates towards the end of the summer season when most crops are almost mature.Most farmers interviewed did not consider it necessary to weed it out as they wereof the opinion that it had no chance to compete with the crop. Both species ofRichardia were observed and reported to germinate throughout the summer andconsequently those plants which grow towards the end of the summer season arenever removed. Tribulis terrestris was observed and reported only in Masvingowhere it is on the increase.

Another observation was that over 40 per cent of the farmers' fields visitedhad weeds and crops of the same size before the first weeding was done. This is toolate to remove weeds as the detrimental effects of weeds on crop growth and finalyield would have already occurred. Most farmers tended not to remove weeds

Table IITHE MOST DIFFICULT WEEDS TO CONTROL IN EACH PROVINCE*

Manicaland Mashonakmd Central Mashonaland East Mashonaland West

Acanthospermum hispidum 18Eleusine indica 12Striga asialica 10Striga spp. 10Commelina benghalensis 8Amaranthus hybridus 6Amaranthus spp. 6Cynodon dactylon 4Cleome monophylla 4Bidens pilosa 3Richardia scabra 3Nicandra physalodes 3Euphorbia helerophylla 2Cyperus esculentus 2Urochloa panicoides 1Hibiscus meeusei 1

Commelina benghalensisAcanthospermum hispidumEleusine indicaStriga asialicaStriga spp.Cyperus esculentusRichardia scabraPortulaca oleraceaCyperus rotundusAmaranthus hybridusPanicum maximumRhynchelytrum repensTagetes minutaRottboellia cochinchinensisNicandra physalodesConyza floribundaDigitaria spp.Xanthium strumarium

19 Acanthospermum hispidum 1518 Eleusine indica 1217 Commelina benghalensis 1012 Cyperus esculentus 811 Rottboellia cochinchinensis 89 Richardia scabra 78 Striga spp. 73 Striga asialica 63 Amaranthus hybridus 33 Galinsoga parviflora 22 Bidens pilosa 22 Urochloa panicoides 22 Digitaria spp. 22 Eragrostis aspera 11111

Acanthospermum hispidum 14Commeline benghalensis 10Cyperus esculentus 9Cyperus rotundus 9Eleusine indica 8Amaranthus hybridus 5Rottboellia conchinchinensis 5Tagetes minuta 5Galinsoga parviflora 3Euphorbia heterophylla 3Panicum maximum 3Ipomoea plebeia 2Rhynchelytrum repens 2Urochloa panicoides 2Bidens pilosa 2

mmDwc<m<

Midlands Matabeleland North and South Masvingo

Acanthospermum hispidum 25Striga asiatica 23Striga spp. 22Eleusine indica 21Commeline benghalensis 20Tagetes minute 11Cyperus esculentus 10Portulaca oleracea 5Amaranthus hybridus 5Bidens pilosa 5Cyperus rotundus 3Cynodon dactylon 3Rhynchelytrum repens 3Urochloa panicoides 1Galinsoga parviflora 1Nicandra physalodes 1

Acanthospermum hispidum 18Bidens pilosa 14Eleusine indica 13Commelina benghalensis 12Cyperus esculentus 10Tagetes minuta 7Portulaca oleracea 5Amaranthus hybridus 4Euphorbia heterophylla 4Nicandra physalodes 3Rottboellia cochinchinensis 2Urochloa panicoides 2Cynodon dactylon 1

Acanthospermum hispidum 21E l i idi 20

pEleusine indica•Sfn̂ a asiatica

g ppCommelina benghalensisAmaranthus hybridusRichardia scabraBidens pilosaTagetes minutaCyperus esculentusCyperus rotundusEuphorbia heterophyllaTribulus terrestrisHibiscus meeusei

201818161088176522

* The most difficult weeds to control are at the top of the list and the least difficult at the bottom of each list. The numbers indicate the number oftimes when the weed was reported and observed to be a difficult weed to control.

OI

zQm

•176 WEED SURVEY

once their crops had reached the flowering stage. The only exception was with thecotton crop where weeds would down-grade the crop if the lint was contaminatedwith trash from weeds.

Of the thirteen most aggressive and difficult to control weeds, eight werebroadleaf and five were narrowleaf (Table III). From the five narrowleaf weedsthree were grasses and two were sedges.

Table III

THE THIRTEEN MOST AGGRESSIVE AND DIFFICULT WEEDS TOCONTROL, CLASSIFIED ON FAMILY AND LEAF BASIS

Weed

Acanthospermum hispidumAmaranthus hybridusBidens pilosaCommelina benghalensisCynodon daciylonCyperus esculentusCyperus rotundusEleusine indicaIpomoea plebeiaRichardia scabraRottboellia cochinchinensisStriga asiaticaTagetes minuta

Family

AsteraceaeAmaranthaceaeAsteraceaeCommelinaceaePoaceaeCyperaceaeCyperaceaePoaceaeConvolvulaceaeRubiaceaePoaceaeScophulariaceaeAsteraceae

Narrowleaf

NoNoNoNoYesYesYesYesNoNoYesNoNo

Broadleaf

YesYesYesYesNoNoNoNoYesYesNoYesYes

During the field survey it was noticed that most areas in the SSNC farmingsector had sandy soils associated with A. hispidum. This observation agrees withThomas (1970) and Drummond (1984). The weed was observed to have greaterplasticity, i.e. it grew up to about 1 m tall, branched profusely and produced manyseeds in moist, fertile soils, and in relatively dry, less fertile soils it grew to less than30 cm tall and yet still produced at least small quantities of seeds. It spread easilyby the attachment of its barbs to human clothes or animal fur (Drummond,1984). This is probably why it has been found in all the eight provinces and all fiveregions of Zimbabwe. Although there have been no competition studies done onthis weed, it does not seem to be competitive with most crops and is easily shadedin tall crops such as maize and cotton.

A. hispidum grew on almost all soil types and produced new generationsirregularly. This is probably why most farmers regarded it as the most aggressiveand difficult weed to control.

Thomas (1970), Budd (1975) and Chivinge (1983) reported A. hispidumtobe among the top most aggressive and difficult weeds to control in the LSCfarming sector. This clearly shows that this weed is a problem in all the arable

O. A. CHIVINGE 177

farming sectors of Zimbabwe, the problem being probably more pronounced inlight soils.

E. indica produces numerous small seeds which germinate quite early in thecrop-growing season and plants have a vigorous and extensive root system (Holmetal., 1977). Field observation during the survey revealed that once the weed wasmore than one and a half months old from the time of emergence, it became verydifficult to remove manually or mechanically. It is, therefore, not surprising tofind this weed among the top most aggressive and difficult weeds to control inmost provinces and ecological regions. Budd (1975) found it to be the mostaggressive weed and the second most difficult weed to control in the LSC farmingsector. This shows that E. indica grows in all ecological regions and soil typesfound in Zimbabwe. Consequently it is a problem in all the arable farming sectorsof Zimbabwe.

C. benghalensis produces both aerial and subterranean seeds and alsoreproduces vegetatively. If the weed is removed by hand or mechanically, stemsbreak off and root at the nodes, producing new plants. Thus, weeding may indirectlymultiply the plant. Budd (1975) and Chivinge (1983) found C. benghalensis to beamong the top seven most aggressive weeds in the LSC farming sector. Fieldobservations by the author revealed that if C. benghalensis was allowed to grow infertilized soils it grew quickly and luxuriantly and easily out-grew the crop plants.Control at that stage would be difficult except probably by the use of chemicals. Thismay explain why respondents reported it among the most aggressive and difficultweeds to control.

Other aggressive weeds such as S. asiatica, Striga spp., A. hybridus, T. minuta,B. pilosa and R. cochinchinensis are allowed to produce seed by the farmers asmost SSNC farmers do not remove weeds once their crops have passed theflowering stage. The result is that these weeds come up in larger numbers in thesubsequent seasons, as more seeds will have been added to those already in theseed bank. If more seed production is coupled with weed seed dormancy then theproblem is worsened. For example, Striga spp. produces 400-500 seeds percapsule and seeds remain viable for 15-20 years (Ramaiah etal., 1983).

While Striga spp. were among the top most aggressive and difficult weeds tocontrol in the SSNC farming sector, both Thomas (1970) and Budd (1975) donot even mention them in the LSC farming sector surveys. This clearly shows adifference in crop and weed management systems in the two farming sectors. Thefact that Striga spp. were more troublesome only in certain provinces is areflection of the extent to which they grow maize, pearl millet and sorghum crops.The Midlands, Masvingo and Mashonaland East had certain areas which grewlarge proportions of these crops. Sorghum and pearl millet were not fertilized withinorganic fertilizers which reduce Striga infestations (Drummond, 1984). Strigaspp. were found growing in patches, and the plants were short and thin. They onlybecame conspicuous after flowering. This makes their control before floweringdifficult.

Budd (1975) and Chivinge (1983) reported R. cochinchinensis to be thesecond most aggressive and the second most difficult weed to control in the LSCfarming sector. However, in this survey it was a problem only in MashonalandEast. The main reason why it is not a big problem in the SSNC farming sector isthat it grows well in heavy soils (Thomas, 1970). There are less heavy soils in the

178 WEED SURVEY

SSNC and more in the LSC farming sector. The fact that it was not evenmentioned in Masvingo and Manicaland is because of unfavourable low rainfalland low temperatures, respectively, as suggested by Thomas (1970). Weedmanagement throughout the whole year, rather than weed control during the firsthalf of the summer, would reduce the numbers of these weeds. The fact that themost aggressive weeds are similar to a certain extent in all the provinces andecological regions suggests that the farming practices used by the SSNC farmingsector are not very different, irrespective of the soil type, amount of rainfall orother climatic factors. Early germination and plasticity are attributes of highlycompetitive plants (Aldrich, 1984) and these characteristics are found in most ofthe weeds listed as aggressive and difficult to control in this survey. Some of theseweeds, such as A. hispidum, have irregular germination which makes controldifficult because they germinate throughout the whole year. Most SSNC farmerscannot weed when it is too wet because they either hand-hoe or cultivate withox-drawn implements. The result is weeds with big root systems which aredifficult to remove by these methods of weeding.

CONCLUSIONS

The most aggressive and difficult-to-control weeds in the SSNC farming sectorhave been identified in all the administrative and ecological regions. The weedswere common across all the ecological regions and were mostly broadleaf weeds,three grasses and two sedges. Weeds which have started increasing over the pastfew years are now known and a base for future weed surveys has been established.It should be appreciated that some weeds might be very aggressive or difficult tocontrol only in a particular limited locality.

Acknowledgements

The author is grateful for the assistance provided by the following: Mr A. R.Maclaurin for correcting the manuscript, Mrs J. Tiffin for the map, Mrs E. A.Wormald for typing, Agritex staff and farmers for their co-operation.

References

ALDRICH, R. J. 1984 Weed-Crop Ecology: Principles in Weed Management(North Scituate MA, Breton Publishers).

BUDD, G. D. 1975 'A second survey of arable lands of Rhodesia', RhodesiaAgriculturalJournal, LXXIII, 159-60.

CHANCELLOR, R. J. and FROUD-WILLIAMS, R. J. 1984 'A second survey ofcereal weeds in central southern England', Weed Research, LXXIV, 29-36.

CHIVTNGE, O. A. 1983 'A weed survey of arable lands in the commercial sectorof Zimbabwe', Zimbabwe AgriculturalJournal, LXXX, 39-41.

DRUMMOND R. B. 1984 Arable Weeds of Zimbabwe: A Guide to theRecognition of the More Important Weeds of Crops (Harare, AgriculturalResearch Trust of Zimbabwe).

O. A. CHIVINGE 179

HOLM, L. G., PLUCKNETT, D. L., PANCHO, J. V. and HERBERGER, J. P. 1977The World's Worst Weeds: Distribution and Biology (Honolulu, HawaiiUniv. Press).

RAMAIAH, K. V., PARKER, C , RAO VASUDEVA, M. J. and MUSSELMAN, L. J.1983 Striga Identification and Control Handbook (Hyderabad,

International Crop Research Institute for the Semi-Arid Tropics).RATTRAY, J. M. and WILD, W. 1955 'Report on the vegetation of the alluvial

basins of the Sabi Valley and adjacent areas', Rhodesia AgriculturalJournal,LII, 484-501.

SOANE, C M . and WAISTER, P. D. 1963 'A note on weeds of irrigated lands inthe Sabi valley', Journal of Agricultural Research, I, 47-52.

THOMAS, P. E. L. 1970 'A survey of arable lands of Rhodesia', RhodesiaAgriculturalJournal, LXVII, 34-5.