Annual report

Material Information

Annual report
Portion of title:
ATIP annual report
Agricultural Technology Improvement Project (Botswana)
Botswana -- Dept. of Agricultural Research
Midamerica International Agricultural Consortium
Place of Publication:
Gaborone Botswana
Physical Description:
v. : ; 28 cm.


Subjects / Keywords:
Agriculture -- Research -- Botswana ( lcsh )
Agriculture -- Technology transfer -- Botswana ( lcsh )
Agricultural extension work -- Botswana ( lcsh )
serial ( sobekcm )


General Note:
Description based on: No. 3 (1985)
General Note:
General Note:
Issued by: Dept. of Agricultural Research, Ministry of Agriculture, Botswana, and Mid-America International Agricultural Consortium.
Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
Statement of Responsibility:
Agricultural Technology Improvement Project (ATIP)

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services ( with any additional information they can provide.
Resource Identifier:
001881601 ( ALEPH )
30952470 ( OCLC )
AJV6698 ( NOTIS )

Full Text
MT/01/11 :C[31NIHJ be6l U360130
(b86T U38143IJ39 C061 lgflDnY :aoiU3d) Z UMN J.VOJ3H lYflNNY

PRINTED: 17110i1984
* *********************************

(b). To feed information on the observed tradItional pa ttern of moderate
weed burdens up to the newly established post of Weed Agronomist at
The first is a description/idiagnostic stage activity. The second is part of a general interest in improving linkages with component researchers at Sebele. A third weed control investigation at the Testing Stage is described in Section
Justification : .A wide range of weed levels have been observed on fields and on plots within fields, in the project areas. Plots that are nearly weed free without weeding can be found among weed problem areas. Weed problems can be attributed to one of two sources: first, poor burial of Grass and other weeds at ploughing. and second, fresh emergence of perennial plants and annual seeds. It is hypothesised that the first type of problem is related to weed levels at ploughing, soil moisture at ploughing, ploughing depth, ploughing speed -- a function of traction type and animal number. The second type of problem should be related to ploughing depth and post-plant rainfall. This combination of management and environmental factors could have relevance for improved weed control recommendations. Traditional management factors or modifications of these factors that contribute to weed control may find quicker acceptance than would recommendations that require more hours weeding.
Approach : This investigation includes description/diagnostic monitoring activities. An evaluation of the current range of weed problems will be continued using the weed growth survey initiated in 1983-8-4. This survey is conducted by ATIP technical staff. The survey assesses incidence, stage of growth and percent area covered for important weed species in the Central Agricultural Region. n the assessment, comparisons are made between different cropping patterns, rainfall patterns and periods of the season. Comparisons are also made between the effects of various experimental tillageiplanting systems on levels of weed control. The Sebele Weed Agronomist will contribute to the design and analysis of this survey. REGIONAL PERSFECT VE STUDIES
It was intially planned that regional perspective studies would begin during the 1983-84 season. The general objectives and justification for regional perspective research were presented in ATIP Annual Report No 1.
(1). Regional Exploratory Survey
Objectives : The primary objectives of a regional exploratory survey are.
(a). To identify where in the Central Agricultural Region changes in
farming systems seem to be taking place or have taken place in the
last 10-20 years.
(b). To find out why donkey and tractor traction have been replacing
cattle traction in some villages but not in other villages.
(c). To rank arable production problems and research priorities based on
File: .AAWSSO/Mah.3 7.45 Date: 17/10/84

According to the contract drawn up between the Mid-America International Agricultural Consortium (MIAC) and AID dated 28th June, 1982, ATiP is expected to provide the following (Appendix A, page 3):
(a). Annual Work Plan to be forwarded in September each year.
(b). Annual Progress Report to be produced in September each year.
As outlined elsewhere (ATIP Administrative Note Number 6 dated March 24th, 1983) ATIP have proposed the following:
(a). Because of the nature of the project, the format of which is less
predictable than, for example, an instructional project, it would
be desirable to produce a report at six monthly intervals. Thus it is suggested one is produced in March of each year which can be
incorporated into the obligatory one required in September.
(b). In order to increase the returns from the time invested in
producing the reports it would be desirable that they be produced
in such a way that they can. be distributed to visitors and other
interested individuals outside USAID and GOE. The reports that are
available for general visitors however will not include all the
material in the Annual Report.
This reporting year no six month report was produced in March. This was due to too many other commitments and preoccupation with field work. it remains to be seen whether in fact the March report will be very practical. It is likely that such reports would not be produced before June. However in June a preliminary draft report was produced especially for use by the Mid-Term Evaluation Team. As such it contained little in the way of empirical results because analysis on current work was still being undertaken. The only difference between that and the current report is that the updated version contains more in the way of empirical results and a more refined work plan for the coming year.
File:.AAWS48/Intro 3.1 Date; 17/10/84

I.OVERVIEW ............... ................. ............. (10 pages)...1.1
1.1 PREAMBLE .. ...... .................................. 1 1
1.2 PERSONNEL... ........... ...... ............................. 1.3
1.3 ADMINISTRATION...............................................1.3
1.4 TRAINING........................................................1.3
1.5 PROFESSIONAL WORK... ............ ........ .................1.4
1.5.1 RESEARCH ACTIVITIES........ ........................... 14
1.5.2 OTHER PROFESSIONAL ACTIVITIES .........................1.8
1.6 WORK PLAN PROPOSED......................................1.8
1.7 PAPERS ... .... ..... ......... ................................1.8
2.PERSONNEL.......................... ...................... (4 pages)...2.1
2.1 PROFESSIONAL STAFF.............................................2.1
2.2 SUPPORT STAFF.......... ........ .. .......................2.3
2.3 CONSULTANCIES ........ ... ... .... .........................2.4
3.ADMINISTRATION.................. ......................... (7 pages)...3.1
3.1. TEAM LEADER.... .. .............. .........................3.1
3.2' DEPUTY TEAM LEADER.... ...... ........ ...................3.3
3.3 OTHER ADMINISTRATIVE TASKS .....................................3.4
3.4 THE MID-TERM EVALUATION TEAM .................................3.4
. 3.4.1 METHODOLOGICAL ISSUES...................................3.4
3.4.2 IMPLEMENTATION ISSUES.................................. 3.5
3.4.4 INSTITUTIONALISATION ISSUES ..........................3.6
3.4.5 PERSONAL DEVELOPMENT ISSUES ...........................3.6
4.TRAINING .. .. ......................... .........(4 pages)...4.1
4.1 FORMAL TRAINING....................... .....................4.1
4.1.1 LONG TERM TRAINING............... ......................4.1
4.1.2 SHORT-TERM TRAINING.... .... ..........................4.3
4.2 ON-- B TRAINING................. ................... 4.4
5.TEAM LEADER, SEBELE. ..... ... ................... (3 pages)...5.1
5.1.1 RESEARCH ACTIVITIES ....................................5.1
5.1.2 OTHER PROFESSIONAL ACTIVITIES .........................5.3
5.2 WORK.PLANNED, OCT.'84 SEPT.'85 ................................5.3
6. RESEARCH EXTENSION LINKAGE, GABORONE.... ............(8 pages)...6.1
6.1.1 RESEARCH ACTIVITIES. ..................................6.1
6.1.2 OTHER PROFESSIONAL ACTIVITIES .........................6.1 SURVEY AND PROMOTION OF RESEARCH EXTENSION COMMUNICATIONS............................... 6.1 IN COUNTRY TRAINING ..........................6.3 FSR ACTIVITY COORDINATION._....... ........6.5 OTHER COMMITTEE ACTIVITY............. .......6.5 VISITS .......................................6.6
File:.AAWS48/Intro S.2 Date: 17110184

KNOWLEDGE ....................................6.7 LOW EXTENSION WORKER MORALE ..................6.7
6.2 WORK PLANNED, OCT.'84 SEPT.'85 .............................6.8
7. MAHALAPYE FARIING SYSTEMS TEAK .........................(61 pages)...7.1
7.1 REPORT OF ACTIVITIES, OCT.'83 SEPT.'84 .....................7.1
7.1.1 DESCRIPTION AND DIAGNOSTIC STAGE...... ............. 7.1 WHOLE FARM STUDIES...........................7.1
(1) Multiple visit resource use survey..........7.2
(2) Technical monitoring.........................7.7
(3) Inventory surveys..... .....................7.9 "NEW VILLAGE" DESCRIPTION AND DIAGNOSIS......7.14 COWPEA BASELINE STUDY.......................7.17 REGIONAL PERSPECTIVE STUDY ...................7.21 DESCRIPTIVE/DIAGNOSTIC WORK NOT MENTIONED IN THE 1983-84 WORK PLAN. .......................7.22
(1) Cropping plans survey .......................7.22
(2) Institutions, services and infrastructure...7.22
(3) Weed growth survey.... ..................... 7.23
(4) Soillroot profile survey.....................7.25
7.1.2 DESIGN STAGE................. . .....................7.25 LOCAL SORGHUM GERMPLASM EVALUATION...........7.25 POST-EMERGENCE HARROWING......... .............7.26 RIDGE PLOUGHING. .. .- ... ..... ............ 7.26 EFFECTS OF EARLY TILLAGE ON TRADITIONAL BROADCAST/PLOUGHING.... ..... ..............7.26 DESIGN ,WORK NOT MENTIONED IN THE 1983-84 WORK PLAN ......... .... ........ ...................7.27
(1) Dual purpose (grain/hay) evaluations in Tswana cowpeas ............................7.27
(2) Bird Scaring scheme .........................7.29
(3) On-farm experimentation methodologies study.7.30 7.1.3 TESTING STAGE..... ...............................7.30 EFFECTIVENESS OF SOLE PLOUGHING..............7.30 DRAUGHT TEAM .MANAGEMENT DESIGN SCHEME........7.31 EVALUATION OF NITROGEN (N) AND PHOSPHOROUS
(P) BENEFITS .................................. 7.32 COWPEA AND MIXED CROPPING COMPARISONS........7.33 EVALUATION OF WEED CONTROL METHODS AND BENEFITS.... ..... ........ ...... .........7.34 LANDS AREA VEGETABLE PLOT........ .......... 7.35 FURTHER INVESTIGATION OF THE MODIFIED EFSAIP SINGLE PLOUGHING/PLANTING OPERATION, ........7.35 SEED TREATMENT...... ......................... 7.35
7.1.4 DATA ANALYSIS AND-REPORT PREPARATION ..................7.36
7.1.5 FARMER WORKSHOP................. ........ ... .. ..... 7.36
7.2 WORK PLANNED, OCT,;'84 SEPT.'85 ...... ......... .. .........7.37
7.2.1 DESCRIPTIVE AND DIAGNOSTIC STAGE....................... 7.38 WHOLE FARM STUDIES .........................7.38
(1) Multiple visit resource use survey..........7.38
(2) Economic analysis of MVRU data.............. 7.40
(3) Activity survey .............................7.40
File:.AAWS48/Intro S.3 Date: 17/10/84

(4) Inventory surveys........................... 7.41
(5) Whole field monitoring ...................... 7.42
(6) Livestock practices survey .................. 7.43 CROP GROWTH RESPONSE STUDIES ................. 7.43
(1) Plot monitoring ... .........................7.43
(2) Weed control investigations................. 7.44 REGIONAL PERSPECTIVE STUDY ...................7.45
(1) Regional exploratory survey..................7.45
(2) Quarterly price survey ......................7.46
(3) ALDEP participant survey ....................7.47 FARMER'DECISION ANALYSIS STUDIES............. 7.47
(1) Decision unit/management information (DUMI) study ..... .... ... ....... .............7.47
(2) Farmer terminology study ....................7.47
7.2.2 DESIGN STAGE...... ......................................7.48
(1) Draught management ..........................7.51
(2) Intensive production plots..................7.52
(3) Undersowing trial.............. ............7.54
7.2.3 TESTING STAGE..... ...................................7.55 IMPROVED TILLAGE/PLANTING SYSTEMS INVESTIGATIONS ......................................7.55
(1) Tillagelplanting scheme trial............... 7.55
(2) Alternative planting methods trial..........7.55 INDIGENOUS PRACTICES INVESTIGATIONS..........7.56
(1) Replant strategy. .... .................... 7.56
(2) Cwpea leaf harvesting strategy............. 7.57 STEPS IN WEED CONTROL ........................7.58 CROPPING SEED COMPARISONS UNDER BROADCAST PLOUGHiNG..... ............. ..................7.58
S DEMONSTRATION... ............................7.60
7.2 4.2 FARMER WORKSHOPS. .............................7.60
7.2.5 LINKAGES WITIH SEBELE RESEARCHERS ......................7.61
8. FRANCISTOWN FARMING SYSTEMS TEAM... ..................(48.pages)...8.1
8.1 REPORT OF ACTIVITIES, QCT '83 SEPT'84.'4 ............. 8.1
8.1.1 DESCRIPTION AND DIAGNOSTIC STAGE. ..................8.2
(1) Exploratory survey.~....................... 8.2
(2) Sample frame survey .........................8.5 AGRONOMY........... ......... .. ...... ......... 8.7 ECONOMICS ......................................8.13
(1) Multiple visit resource use................. 8.13
(2) Village infrastructure survey............... 8.16
(3) Cowpea survey............. ..................8.16 ANIMAL HUSEANDRY .............................. 8.18
(1) Livestock practices survey ..................8.18
(2) Plough condition survey .....................8.20
8.1.2 DESIGN STAGE.... ......................................8.20 MAXIMUM PRODUCTION TRIALS.....................8.20
8.1 2 2 STEPS IN TECHNOLOGY EXPERIMENTS.............. 8.23 STAND ESTABLISHMENT .........................8.25
File:.AAWS48/Intro S.4- Date: 17110184 DEPTH OF PLOUGHING ........................... 8.26 OTHER DESIGN WORK NOT IN THE 1983-84 WORK P LAN .........................................8 .26
(1) Late planted crops for forage............... 8.26
8.1.3 TESTING STAGE .........................................8.26 AGRONOMY .....................................8.26 ANIMAL HUSBANDRY .............................8.27
(1) Supplementary mineral-mix for farm animals..8.27 8.1.4 DISSEMINATION STAGE. ....................... ...........8.29
8.1.5 METHODOLOGY DEVELOPMENT EXPERIMENTS ...................8.29 HARVEST DATA COLLECTION ......................8.29
8.2 WORK PLANNED, OCT.'84 SEPT.'85 .............................8.32
8.2.1 DESCRIPTIVE AND DIAGNOSTIC STAGE ......................8.32 ARABLE ACTIVITIES ............................8.32
(1) Economic and technical monitoring...........8.32
(2) Maximum. production plots ....................8.34
(3) Agronomic baseline survey ...................8.36 LIVESTOCK ACTIVITIES .........................8.37
(1) Economic and technical monitoring...........8.37 Non agricultural activities .................. 8.38 Agricultural markets and infrastructure......8.38 .Institutional arrangements...................8.39
8.2.2 DESIGN STAGE .............................. ..............8.39 PLANTING STRATEGY ............................8.39
(1.) Planting strategy ...........................8.39
(2) Draught management for early ploughing......8.40 ANIMAL HUSBANDRY .............................8.41
(1) Improved harnesses and yokes for draught an imals .....................................8 .41
(2) Adequate drinking water for working animals.8.42
(3) Crop residue and forage preservation........8.42 ECONOMICS. ................... .................8.43
8.2.2 TESTING STAGE .... ....................................8.43 AGRONOMY ........................................8.43
(1) Steps in technology .........................8.43
(2) Testing of cowpea varieties .................8.43
(4) Testing of EFSAIP planting equipment........8.43 ANIMAL HUSBANDRY..................... .......8.44
(1) Supplementary mineral-mix for farm animals..8.44 ECONOMICS............... .................... 8.45
8.2.4 DISSEMINATION STAGE ...................................8.45
. ANIMAL HUSBANDRY...............................8.45
(1) Prepacked mineral-mix .......................8.45
(2).Dissemination of DAFS and veterinary progilmme information.......................8.46
8. .5 OTHER PROFESSIONAL ACTIVITIES ........................8.46
- , METHODOLOGY TRIALS ...........................8.46 LINKAGES WITH ATIP MAHALAPYE .................8.47
8:.2.5.3 LINKAGES WITH ON-STATION RESEARCH AT SEBELE..8.47 8.2,5.4 LINKAGES WITH EXTENSION ......................8.48 LINKAGES WITH ALDEP ..........................8.48
9. MEETINGS, VISITORS AND PAPERS ............................ (5 pages)..9.1
9.1 MEETINGS AND TRIPS ,... ....... ..............................9.1
9 .2 V ISITORS ........ .. ...... ... ..... .......... ...... 9... 9 1
File:.AAWS481Intro .5 5 Date: 17110184

9.3 POLICY ON PAPERS WRITTEN BY ATIP STAFF ....................... 9.2
9.3.1 EXTERNALLY PUBLISHED PAPERS ........................... 9.2
9.3.2 RESEARCH REPORTS ...................................... 9.2
9.3.3 WORKING REPORTS ....................................... 9.2
9.3.4 MISCELLANEOUS PAPERS .................................. 9.2
9.3.5 PROGRESS REPORTS ...................................... 9.2
9.3.6 DOCUMENTS REQUIRED BY USAID ........................... 9.3
9.4 PAPERS .... ........................................93
9.4.1 EXTERNALLY PUBLISHED PAPERS ........................... 9.3
9.4.2 WORKING PAPERS ........................................ 9.3
9.4.3 MISCELLANEOUS PAPERS .................................. 9.3
9.4.4 PROGRESS REPORTS ...................................... 9.4 GABORONE ..................................... 9.4
9.4.4 .2 MAHALAPYE .................................... 9.5 FRANCISTOWN ................................... 9.5
9.3.4 REPORTING DOCUMENTS ................................... 9.5
File:.AAWS48IIntro -.6- Date: 17/10/84

SEASON.... ........... ... ... .............. ... ... ...........1.6
SEASON.................................................... ....1.7
1985........................ ..... .. .................... ...1.9
SEPTEMBER 1985. ................................................1.10
2.1 PERSONNEL ON THE ATIP PROJECT .................................2.2
4.1 PARTICIPANT TRAINING ..........................................4.1
MAHALAPYE, 1983 84 ...................... ...................7.4
JUNE 1984. .................................................... 7.5
1983 84 ................. ....................... ............7 .9
MAHALAPYE AREA, 1983 84 ....................................7.10
7.5 FIELDS AND PLOUGHS, MAHALAPYE AREA, 1983 84 .................7.12
- 8 4 ... . . . . . . . . ... . . . . . . . . . . . . . . . . . . .7 .1 3
MAHALAPYE AREA, 1983 84 .................. ..................7.15
SEASON, MAHALAPYE AREA, 1983 84..............................7.23
MAKWATE VILLAGE, 1983 84 ....................................7.25
MAHALAPYE 1983 84 ..........................................7.28
TRADITIONAL CHECKS, MAKWATE 1983 84 ........................7.31
PLOUGHING COMPARISON, MAKWATE, 1983 84 ......................7.33
SYSTEM, MAKWATE, 1983 84 ....................................7.34
AREA 1983 84 ....................... .......................8 .6
File:.AAWS48/Intro S.7 Date: 17/10184

SEED SAMPLES, FRANCISTOWN AREA, 1983.........................8.9
AREA, 1983 84................... .. ........ .. .... ...... ...8.19
TRIALS", MATHANGWANE AND HARAPONG, 1983 84..... .. ...... ..8.Z2
MATHANGWANE, 1983.- 84.- ........ .3-....-.... .8.23
TUTUME DISTRICT, 1983 -84 ....... ... .. ....8.24
File:.AAW48/Intro S.8 Date: 17/10/84

1.1 THE FARMING SYSTEMS APPROACH .......... .. .........1.2
Fi1e:.AAWS48Intro S.9 Date: 17/10/84

Many of the points mentioned in the overview are discussed in the more detailed chapters that follow.
The major purpose of the ATIP project is to improve the capacity of the Ministry of Agriculture's (MOA) research and extension programmes to develop and effectively extend farming system recommendations relevant to the needs of the small (limited- resource) farmer. In the project design it was envisaged that ATIP would contribute to this purpose in three ways:
(a). Help strengthen the experiment station based research of the
Department of Agricultural Research (DAR) and strengthen the links
between such research and farmers by implementing a Farming Systems
Approach to Research (FSAR), with teams based at Mahalapye and
(b). Help strengthen the linkage between research workers in DAR and
extension personnel in the Department of Agricultural Field
Services (DAFS) through the appointment of a Research Extension
Liaison Officer (RELO).
(c). Help set up a commercial seed production system to ensure that
adequate supplies of needed seed are available for distribution to
Batswana farmers.
In terms of the above, this report only includes consideration of (a) and
(b) above. Implementation of (c) has been postponed as a result of discussions between USAID/Botswana (USAID/E) and the Government of Botswana (GOB).
A number of attributes of FSAR helped focus the direction of ATIF's work. Since there is a great deal of confusion as to what is called Farming Systems Research (FSR), a conceptual outline of the ATIP approach is given in Figure 1.1. ATIP researchers prefer to call it a Farming Systems Approach to Research (FSAR) rather than the more popular term FSR. ATIP's FSAR program is aimed at improving the welfare of farming families through increasing their productivity. Two complementary approaches can be followed:
(a). Farming Systems Research (FSR) -- involving the development of
relevant improved technologies and their dissemination via the
extension service.
(b). Farming Systems Perspective (FSF) -- involving influencing the
development of relevant policies and support systems.
Initially ATIP's emphasis has been mainly on FSR although hopefully later in the project increasing emphasis will be placed on FSP. ATIP researchers
File:. AAWS48/Over 1 Date: 17/10184

F = Farmer
H flnae
1. Descrictive I Dia1xostic _________________________Current farming system Sucoort systems
~hypshesi fgrm~iatcn) d poli
2 einUse body of ,. Eluer imerti
izom '. ,,\
3. Teitino ---- ---- -- - -- -
FM IN --- - - - - -
-A. dismU~~ di f i ad f.&einn -------- ----- -----File:.AA348EIval 151"1 Ctat "t84

believe a successful FSR programme must include involving the farmer (the
consumer of the improved technologies); tapping the pool of knowledge of farmers; recognising the heterogeneous nature of the farming community;
using an interdisciplinary approach; exploiting complementary and
supplementary relationships in the farming system, and having a dynamic and
iterative research approach which is complementary with experiment station
based research.
Apart from a lack of a counterpart for Norman there was a full complement of ATIP senior level staff through most of the reporting period. Although
there was no anthropology/sociological input by GOB during the reporting
period, an agreement has been reached for a person to be seconded to the project full time for a period of one year starting in July or August. In
the March/April period the INTSORMIL CRSP agronomists (Carter and
Youngqulst) arrived and are starting to interact with ATIP personnel.
By the end of the reporting period most of the problems with respect to
support staff had been resolved. Enumerator positions have been
established by GOB for ATIP, while clerical staff have been posted to
Francistown and Mahalapye. The major remaining problem is with respect to
the T4 positions. Two more have been appointed, leaving two still to be recruited. With reference to the eight approved GA5/6 positions, six more persons are supposed to- be provided -nd hopefully will be appointed in the
next couple of weeks.
Three consultancies were implemented during the year in areas as diverse as
the biological fixation of nitrogen, in-service training programmes for
-- extension--staff, and instruetlYI .ATIP staff aid counterparts on the use of
the three Apple microcomputers that have now been purchased .with project
Administrative matters continue to take a great deal of time on the part of the Team and Deputy Team Leaders. However many of the problems unresolved at the beginning of the reporting period have been, or are in the process of being, resolved. Ways are being sought to streamline some of these day-to-day routine administrative tasks, particularly those of a reporting nature, through setting up files on microcomputer disks. It is hoped that over time these can increasingly be maintained by clerical .staff in ATIP.
In July a Mid-Term Evaluation Team of five individuals spent a couple of weeks looking at the project. A report has been produced which has been submitted to the GOB.
Th i individuals sent on long-term training -- two MS and four BS degrees --during the last reporting period were- still away at the end of the current reporting period. No more have been sent during the current reporting period since 16 of the 22 person .years initially to be funded under the project have already been committed. ATIP personnel, as indicated in the 1983-84 Annual Report, are very concerned about the
File:.AAWS48/Over 1.3 Date: 17/10/84

current shortage of funds for long-term training. However, the USAID
Director agreed that the issue of training could be brought to the attention of the Mid-Term Evaluation Team. It is understood they recommended further funding of long-term training. Therefore it is hoprd
that this recommendation will be implemented.
With reference to short-term training four individuals we:e sent on courses
out of the country while 11 were instructed on the use of the Apple
microcomputers that have been purchased by ATIP.
Rainfall figures in Table 1.1 indicate that the two years of ATIP existence have coincided with two bad years in terms of rainfall for arable agriculture. Thus ATIP has been faced with another difficult year as far as work of 'a technical nature is concerned. The figures do show the inadvisability of looking at total annual figures only. There was, for example, during this current reporting period, considerable rainfall late in the season.- However these obviously were too late to be of much use, although they did resultiI a :proloigation' of' the poor' harvest season by encouraging some of the tillers on, cereal crops to head and produce grain.
*MONTH ... ..: .. .
MON ----- -.----- ------------- r--------30 YEAR 1982-'83 1813-84 310 YEAR 1982-83. 1983-84
July 2.1 0.0 1.4 0.5 0.0 0.0
August 0.6 0.0 2.1 0.8 4.1 12.0
September 9.7 2.4 0.2 7.9 0.0 0.0
October 30.7 96.0 17.4 24.3 62.7 89.0
November :' 67.2 48.9 : 1.8 61.7 24,.4 122.5
December 93.9 -85.9 58.5 10i:6 38.7 84.2
January 85.4 26.0 'i.0 98. 60. 4.8
February '91: 7 15.4 24.2 83 5. 41.7 20.4
March 48. 43.1 257.2 46.1 42,.4 145.5
April 31.2, 14.9 25,9 108.3 17.2
May '12.3 1 .9 b 10.9 6.1 0.0
June 4.3 3.0 0.0 2.2 0.0 0.0
------------------ ------------- ---- ---TOTAL 477.5 307.5 515.2 464.1 3889.. 495.6
The two seasons have brought home to ATIP personnel three observations:
(a).. That because ~df the harsh climatic environment there is little
flexibility in the timing :of farm operation'.. Thus interventions
have to address the issue of breaking bottlenecks r other than
S.ile::AAWS48Ovet 1.4 Date: 17/10/84

exploiting flexibility in the farming system.
(b). That farmers, because of great variability in the inter- and
intra-annual distribution of rains,. are faced with a "decision
tree" type, of approach to farming. in other words decisions on..
famn atr vary according to how, te cr ppig#g eyo ~$ps.
(c). That there is not a great deal in the way of technologies on the
shelf that will give. reliable results in every year. Basically
this is because of the tremendous variability in the cropping years
referred to above.
Consequently, unlike farming systems projects in many countries, ATIP is having to put a great deal more emphasis on descriptive/diagnostic work in order to properly understand, not only the human but also the technical environment, in order to introduce strategies that have a reasonable chance of adoption, and'also in order to be able to feed back priorities to scientists on the experiment station. In a more equable climatic environment such emphasis on describing and diagnosing the technical environment would not need to receive so much concentration.
ATIP is also starting to deviate significantly from farming systems projects in other countries in another respect. This is the relative degree of emphasis that ATIP staff are beginning to place on researcher managed and implemented (RM,RD type of work. Once again this arises from the harsh nature of the climatic environment. Working at a level involving more management or implementation by the farmer involves more cooperation on his/her part. This is particularly difficult to achieve during drought years and tends to give highly discounted results, even if in fact they produce anything at all. ATIP has placed a great deal of emphasis on farmers' cooperation for the last two years and therefore have had very few technical results that look promising. Also, because of this, a dialogue with farmers with respect to results has been very limited. By increasing the relative degree of researcher managed and researcher implemented type work, it will be possible to test the feasibility .of certain strategies, while. at the same time it will open up the possibility of dialogue with the farmer, who can see results with his/her own eyes on his/her own field. Once again in a more equable climate where timing is not so critically important, it would not be necessary nor even desirable to place much emphasis on research managed and implemented (Rt4,RD) type work.
The research activities of the ATIP teams ,located in Mahalapye and Francistown are summarised in Tables 1.2 and 1.3. .The results are discussed in much more detail in Chapters 7 and '. However it is interesting to note that, although there is a certain. parallelism in the work of the two teams, they do have their own distinct research identities. This brings up an issue with 'respect to tATIP work. Because of the very uncertain nature of the "environment and the difficulty of designing strategies it does appear justified that the -research agendas in the two places, should be able to differ' according to the needs expressed by the farmers, and according .to the perceptions of the individual researchers. Obviously promising results in one .area could be tested-in another area at a later date. However .at times it-.may appear that there is little coordination between the. work of the two teams. Efforts are being made to
File:.AAWS48/Over 1.5,- Date: 171/10/84

tIlmtI W Tm -rAMID:
DIAGNDSTIC Whole Farm Studies:
MVRU Oct 83-Sep 84 Implemented as planned; 27 farmers
Technical Monitoring Qct 83-Jul 84 Completed on 47 farmer's fields
Whole field Honitoring --------- Replaced by plot monitoring
Special Plot monitoring -------------- Replaced by plot monitoring
New Village Work, May 84-Jul 84 Exploratory mostly complete;
Sample Frate Census dropped
CowpeaBaseline Study May 84-Sep 84 Survey completed 51 farmers
Regional Perspective Study Jul 84-Sep 84 Not started
Cropping Plans Oct 83-Nov 54 Completed; 45 farmers
Institutions, Services, and
Infrastructure Jan 83-tar 84 Completed; villages
Inventory Surveys Feb 84-Hat 84 Completed; st least 25 farmers
Weed Growth Feb 84-May 84 Completed; 59 plot situations
Soil/Root Profiles Apr 84-Jan 84: Completed; .15 fields
.....--------------------------- ----------- ----------------------- ---------------------------DESIGN TRIALS-PLANNED:
-Local Sorghum Geraplasm Dec 83-Jul 84 Completed as planned
Post-Emergence Harrowing 711.2 -Dec 83 Arranged, not implemented
Ridge Ploughing --------- Not arranged, equipment arrived late
Effects of Early Tillage Nov 83-Jun 84 Completed; 2'farirs
Bird Scaring 71.2.3(2) .Feb 84-Nay 84 Completed 4 sites
On-Fara Experimental
Methodology Study 7 .513) Jn 84 Completed 5 plots
Dual Purpose Cowpeas 7 1 51) Dec 83-Jun 84 Completed
-------------------------------- .... ...- -- -- ------- ------.3 -----TESTING TRIALS-PLANNED: .
N and P Benefits 7.1 3.3 1o#44Not 8%-&a P: 8 replicattens implemented
N: Not implemented due to drought
Effectiveness of Sole PloughiagT1.3.1 : -Nov 13-Jun-84 Arranged 14 farmers, I Implemented
Draught.Team Management 7.1.1,1 Oct 83-Apr84 Completed; 5 farmers
Cowpea Cropping Comparisons 7.1.3,4 Dec 83-Jua 84 tArranged 30 farters; 16 implemented
Evaluation of Weed Control 7.1..1 5- --------- .. Replaced with aoitoring
Lands Area Vegetable Plot Nov 83-Dec 83 Arranged 1 farmer; not implemented
EFSAIP PloughlPlanter 7.1-317 Nov 83-Dec 83 Arranged3 faers; implemented 1 farmer
Seed Treatent :Dec 83-Jun 84 Arranged 15 fares; 12 implemented
To few results t1 analyse
File: .AAW v- 1.4 Date: 18110184

Exploratory 8.1.1(0) Sep 83-Oct 83 Completed as planned
Sample Frame 8.1.1(2) Oct 83-Nov 83 Completed as planned; 1367 farmers
ultiple Visit Resource Use Oct 83-Sep 84 Initiated with 31 farmers;
Complete data for 26 households
Village Infrastructure 8.1.1.(2) -Feb 84 Completed in project villages
Cowpea Baseline 8;1.1.2(3) Feb 84-Apr 84 Completed with 275 farmers
Plough Condition Feb 84-Apr 84 Completed with 30 farmers
Livestock Practices 8:1.1.3(1) Apr 84-May 84 Completed with 42 farmers
------------------------------------------------------------------------- -------DESIGN TRIALS:
Haximum Production Nov 83-Jun 84 Planted with 3 farmers
2 trials killed by drought
Steps in Technology Nor 83-Jun 84 Planted with3 farmers
2 trials killed:by drought
Stand Establishment Nov 83-Jun 84
Depth of Ploughing 8.1.2 4 Nov 83-Jun84 Plantedwith 3 farnersi
All killedby drought
Late Planted Crops for Forage Apr 84-Jun 84 3 plots each ofi ovpeis and millet;
2 plots of teff
Yield Estimation Methodology Har 84-Jul 84 Completed on 7 plots
------------------- -------------------------------------------------------------------TESTING TRIALS:
Planting Equipment Nov. 83-Jan 84 Completed with one farmer
Cowpea Variety Dec 83-Jun 84 Arranged with 20 fares; planted by 1
Feeding of Mineral-Mis to
Animals Nov 83-Sept 84 Continuing with 13 0c'attle herds;
29 goat herds; and 12 donkey herds
File:.AAVS4810ver 1.7 Date: 18110184

ensure frequent contact between the two teams. For example, it is now planned to implement quarterly meetings of all. ATIP personnel plus some personnel from outside the project itself (eg, Cowpea CRSP, INTSORMIL, EFSAIP personnel, etc.).
With the recent acquisition of two Apple lie microcomputers it is anticipated that processing/writing up link will be made considerably shorter consequently helping to improve the productivity of ATIP personnel. It is anticipated that data entry onto data bases will continue to be the major responsibility of personnel at Sebele while data analysis will be emphasised at Mahalapye and Francistown. Word processing work obviously will be done at all three locations.
The RELO and his counterpart have continued their visits to regions and districts and have attended many of the monthly meetings. They have also emphasised increased contact with the Crop Production Officers in the regions. In order to obtain a better feel for the problems and perceptions of extensionri and research personnel a survey was devised and administered to personnel 'i DAFS and. the.Arable Research Division of DAR. The surveys have beei completed and returned and soon will be. analysed on the microcomputer at Sebele;
.ther major thrusts during the reporting period were to organ.ise a review of the current Agrifacts .to determine the accuracy of their content and their readability. Also an effort has been mf~de to organise six monthly meetings of the regional. Crop Production Officers at which contact with DAR personnel will be emphasised. This lack of contact between.research and extension has been perceived as a major-problem. It is hoped that this problem will be ameliorated during such meetings Also the RELO and his counter rt'have been involved with a number of committees'lus playi'a role in co rdiftating FSR activity among the various projects in Botswana.
Adh-ministrative-issues are likely to continue to occupy a substantial amount of time especially on the part of the Team.and Deputy Team Leaders, but it is hoped that now ATIP has been firmly established, relati,'eIy- greater emphasis can be devoted to professional tasks.
The major research activities will continue to revolve around the teams at Mahalapye and Francistown. Summaries of their proposed research programmes are shown in Tables 1.4 and 1.5 but detailed descriptions of their plans are presented in Chapters 7 and 8.
A complete list of papers produced by ATIP since its inception is given in Chapter 9.
File:.AAWS48/Over 1.8 Date: 17/10/84

Whole Farm Studies:
(a) MVRU Survey 12 Households O1ct'84-Sep'85:
(b) Activity Survey 40 Households 1Oct'84-Jun'85!
(c) Inventory Surveys 52 Households flNov'84-Jon'85:
(4) Whole Field Monitoring 40 Farms !0C084-Jun'851
(e) Livestock Practices Survey 50 Households JApr'85-May'8: (f) Cattle Post Visits 5 Posts INo9'84-Ju185:
Crop Growth Response Studies:
(a) Plot monitoring 12 Farms !Oct'84-Jun'85l
(b) Weed Control investigations 50 Plots !Dec'84-Jun'85:
Regional Perspective Study:
(a) Exploratory Survey DAFS OfIicers
10 Villages Oct'84-Feb'851
(b) Quarterly Price Survey 7 Villages INor'34-Sep'8S1
(c) ALDEP Participant Survey 50 Farmers 1may'85-Aug'851
Farmer Decision Studies:
(a) Decision Unit/Management,
Information Study 50 Households !Nor'84-Jun'85:
(b) Farmer Terminology 20 Sources Nov'84-Sep'851
Commercial Steps in Technology 3 Fields ,0:t84-Joa'83
Ridge Ploughing Trial 2 Fields 3lick84-Jun'851
Steps in Weed Control 2 Fields Oct'84-Jun'85
RM :
Replanting 30 Replicat as 84-Jan 85I
Draught Management 2 Teams .Oct'84-Jun'85:
Intensive Produttion Level 1 6 Farms :0ct'8-Jan'8:
Intensive Prodiffton Level 2 3 Farms Oct'84-Jan85
Undersoving- i Replications Oc'84-Jn'831
Tillige Planting Scheme 10 Farms 1Oct'84-JanS85
Crop' ending comparisons ct'84-Jun'851
Cowpea Harvesting 20 Replications INovI84-Joe'85g
Alternative Planting Methods 10 Farmers INov'84-Jua 85
Tillage/Planting Schime 6 Locations :0ct'84-Jun'85:
File:.AAVS4810ver 1.9 Date: 18i10184

Monitoring Arable Activities by Plot 9 Farmers :Oct'84-Sep'851 Monitoring Household Activities 30 Farersa lOct'84-Jun'85;:
Monitoring Environmental Variables Variable :Oct'84-Jun'85:
Stand Establishment by Plot 6 Farmers :-------------- Ocet'84-Jun'85--------------Baseline Survey (Tutume District) 275 Households:Nov'84-Dec'84i
(a) Agronomic Fractices
(b) Animal Husbandry Practices
(c) Resource Endowments
(d) Agricultural Markets
(e) Income Sources
Institutional Studies (Case Studies) Variable :Feb'85-May'85:
(a) Access to Resources
(b) Access to Markets
(c) Transfer Payments
(d) Market Structure for Grain and Smallstock
---------------------------------------------- ------- ---------------------------------TRIALS:
Hazinma Production Trials 3 Farmers ?Oct'84-Jun'85
Yield Estimation Methodology 6 Plots O0ct'84-Jun'851
Livestock Blood Phosphorus Levels 11. Farmers 0;ct'84-Sep'83: M.FI
Planting Strategy 2 Farmers :Oet'84-Ju'85:
Draught Management 6 Farmers :Oct'84-Jun'851
Steps in Technology 3 Farmers :------- Oct'84-Jun'85 ------Harnesses and Yokes 8 Farmers : -------Oct'84-Jan'85------Drinking Water for Draught Animals 3 Farmers .--- Oet*84-Jan'853--Feeding Mineral Supplement to Animals 61 Herds :Oct'84-JunS5S
Planting Equipment Tests 3 Farmers :Oct'84-Jan'+85
Cowpea Variety Trials 30 Farmers ;Oet'84Jun'8 5:
The sample of 30 farmers includes the nine farmers where the agronomic activities are monitored by plot.
For these nine farmers no additional information on arabic activities will be obtained.
File:.AAWS481Over 1.10 Date: 1811 0184

The long-term expatriate staff financed under the MIAC/USAID/GOB agreement
were all present throughout the reporting period. There were no major
problems that interfered with their work. Certainly a great asset has been
that all the ATIP long-term expatriates have had previous overseas
experience enabling them to quickly adjust to working in Botswana. The
housing situation seems to have stablised although the Francistown
technicians are still living in private housing. :However since the private housing is very satisfactory there have been no issues arising from thi decision by the GOB to rent them for a further year.
Towards the end of the reporting period the personnel recruited under the INTSORMIL CRSP (Carter and Youngquist) arrived. Limited support to them is given through the ATIP project. Their remaining,:support costs derive from GOB and the INTSORMIL CRSP. However it is anticipated that close-. collaboration will be developed between the work of these technicians and ATIP personnel.
It is encouraging to note that all four of the technicians whose two year tours of duty finish later this year have expressed.a desire tb extend. Hobbs will be returning for onre. further year and Siebert, Baker and Norman are returning for further two year tours. Continuity of FSAR'work is extremely important especially in a harsh climatic environment such as Botswana.
All counterparts sent away on training during the last reporting period in fact are still away. The first one to return will be Monyatsi with an MS degree in Agricultural Economics in January 1985. By the end of the current reporting period most Pt..f the USAID technicians: were working with replacement counterparts. Siebert had to wait until January 1984 for Luzani who was completing a diploma in BAC, while.Matlho joined Koch in February after an orientation period with APRU at Sebele. Norman had no counterpart for the whole of the reporting period. In the case of Siebert the late arrival of Luzani did interfere with work at the beginning of the cropping year since he was until that time the lone. agronomist in Mahalapye. In the case of Koch it was difficult to initiate w Work with :ATIP Mahalapye until after the arrival of Matlho.
Unfortunately during the whole of the current reporting period no sociologylanthropological input was available to the project. According to the terms of the agreement between USAID and GOB such services were to be provided by GOB. However just before the end of the .reporting period an agreement was reached with GOBI for John Lesetho who is employed in the Rural Sociology Unit of DPS to jointhe project in September,.on a full time basis for one year. In essence this will help rectify the lack of an input over the first two years when anthropological/sociological services were to be provided on a half-time basis. At the end of one year's full time work with the project Lesetlho and other members of the Rural Sociology Unit
File: AAWS48/Pers 2.1 Date'. 17/4i0/84

Hobbs, A.' Agronomist Ph.0 Gaborone Aug. 1982
Norman, D.W. Ag. Eton. Ph.D Sebele Aug. 1782
Siebert, J.C. Agronomist Ph.D Naha I' a pye Bept. tyll.
Baker, D.C Ao. Eton. MS s. Mahaiapye 01 lt 98.2
Heinrich, G. Agronomist PLO. :Frauili-itow-n Aagi 19*84
loth, B. Aniafl St. Ph.D Francistown &to. IYA3
Miller, V. Ag. Eton. Ph.D fraheistown'Aug. 1983
Modiakqotla, E. Agronomist 98 Mahalapye, Aug. 1182 Aug.11193
Monyatsi, T. 1g. Econ. B.;8c Nihalapye Aug. 1.982 Jan.1183
Tjirougo, M. Ag. Eton., DA Mahalapye Sept.1182 Aug.1783
Moreaedi, G. Agronomist DA Gaborone Dec. 1982 Aug.1?83
Masikara, S. Agronomist CA Francistown Sept.1993
Mattho, V. (Hiss) Animal Be. B.Se Frat1cistown Sept-1183
Razolemani, C., Agronomist B.Se Gaborone Sept.1983
Selaka, T. ig. Ec.on. DA rrantistown'sept.19831
Tibone, C. (Miss): Aq. Eton. Ok MahiIapye Sept.1793
Mo phutlng', : N (Miss): Welt Sept.1982 jaft.IiI3
Seretse, L. (Mrs) Sebile, mar, 1983
Makelo, Envaiiator T alhe'i -Shoshang", Stpt.'198f
- 6 In! :.: I .. ".
Moilhokod. g' S h' h
ise, '-Euvzara or Vii et' hos d q- ep
'Eunirit r Tr ainei Nakwatez, ep
'Dira, D -Extent'lon- --.14 Hatoate 0 cf-. 9 i 2
leipelle, W. 11tensfou 14 Shosh*onq Octl
Imopid Driver Fr 'hc:is to wn Se'OtA982
Magotsf, 1.. :Drive:' 1* Nahalipye,' Oct.. 1902
:k6nilidt P.: (Miss)' toi Put ing T iaiiee fie"bi 1'e, .11 Feb,''19 83
M.- "Endaerator 'GA6 Hathanowane act. 1983
:Sibanda,-B (Hiss) Enumerator GA. 6 marapong Oct. 1983
Clifford$' 4. (Mrs) Typist S Francistown Sept.1903
Seleke, K. (Hiss) Typist 84 Mahalaple Dee. 1983
Sibanda'-?. Extension T4,. Mataba June 1994
Notes: The table does not include those thatwere employed as casual labour., Also it,
does not include the BAC students who workvith the project for short period.
File: AAV541/Pers 2. 2 Date,: '13110184

will be available to carry out additional studies if and when they are
required. It would be anticipated in such a case that transport and per
diem costs would be met out of ATIP project funds.
A continuing concern is the apparent lack of incentives for Batswana staff
on the project. Some of the counterparts work long hours in the field,
which is to be expected of professionals doing FSAR type work.- However there seems to be little in the way of a reward system for such work and
for sometimes living under somewhat difficult conditions. It is difficult to see what can be done about resolving the incentive issue but it is a
problem that will need to be addressed if the FSAR is going to be
successfully institutionalised in Botswana and if Batswana are going to be
willing to devote their careers to this type of work.
The clerical situation has improved greatly during the last year. Clerical staff or typists have now-been- posted to both Mahalapye and Francistown.
Hopefully the productivity of these individuals will be improved with the
training they have recently received on the Apple Ile microcomputers (see Section 4 .1(b)) and new typewriters that have been provided. There have
been no problems over the last year with reference to the drivers posted to
Mahalapye and Francistown. A second driver is in the process of being recruited for Francistown to help the animal scIentists who work in both Mahalapye and Francistown.
Also another issue that has been resolved satisfactorily during the last year has been that with reference to the "trainees". Tw of the three trainees initiaiy, employed in the Mahalapye area are still with the project.- Recently they successfully passed examinations for appointment to T5 posts.
Another Issue that has been satisfactorily resolved this year has been approval of eight GA516 enumerator posts for ATIP in the -new financial year starting in April 1984. Until now most of -the interviewing work has been done by temporary employees. The approval of permanent -.posts will enable the project to build up a permanent cadre of well trained. enumerators. In the Francistownr area two of the individuals currently employed on a temporary basis have been offered permanent GA6 positions. The Direltbrate of Personnel is currently filling the other vacancies will be held vacant.
The major continuing problem that the project has had during the last year is with respect to the lack of T4s that were supposed to be seconded to the project from DAFS. The project still only have three T4s with another one due to start work with the project shortly. It is planned that two each will be stationed in Mahalapye and Francistown. This means that there is still a need for two further individuals to enable one person to be posted to each of the six ATIP villages. DAFS have repeatedly indicated that they are willing to let us have the T4s but as yet they still have not materialised It is to be hoped that they will be posted to the project very shortly. Their participation in the project should greatly facilitate improved productivity of project personnel in the field.
File: .AAWS48/Pers 2.3 Date: 17110184

It was very encouraging for the project that the Dean of Agriculture Kansas
State University Dr.J. Dunbar made an Executive Visit to the project in March. -He did a substantial amount of travelling within the country and visited all the villages that are currently involved in ATIP work. There
was opportunity for project members to interact with him and raise various
In addition there were four short-term consultancies. They were as
(a). Dr. George Ham, Head of the Agronomy Department visited the project
at the same time as the Dean of Agriculture from Kansas State
University. In addition to working out details about the proposed
posting of INTSORMIL personnel which is being coordinated through
his department, he consulted with ATIF agronomists on grain
legumes. He also had discussions with the Cowpea CRSP Team. He also gave a seminar on "biological nitrogen fixation by legumes" to a large group of research and extension staff. (27th March 12th
(b). Dr. Robert Johnson who is with the Extension Service at Kansas
State University visited and gave advice on how in-service training
within DAFS could be improved (see Section He also
gave a seminar on his findings. (6th June 3rd July).
(c). Mrs. Sandra Miller wife of one of the USAID technicians gave two
two-week courses on the use of microcomputers for ATIP personnel
(see Section 4.1.2). (1st May 15th June).
(d). Dr. David Rees and Mr. ,John. :Sinclair, who recently finished
assignments wth theDry Land :Farming Research S'cheme (DLFRS) at
Sebele, provided the project with unpaid consultancy work in the
area of soils in Botswana. (Selected days in June and July).
Drs; Dunbar,:Hai, and Johns'on have provided reports which are available. The Slriclair/Rees,: report is bin' ompiled with the hep of ATi 'personnel while the consultancy report of.Miller consists of the teaching materials used in the codises..
File: .AAWS48/Pers 2.4 Date: 17/10/84

The. amount of time spent in administrative tasks of all team members was still considerable during the current reporting period although the greatest burdens continued to be borne by the Team and Deputy Team Leader. The administrative tasks continued to be complicated by having team members located in four different places (Francistown, Mahalapye, Sebele and Gaborone), by ATIP .having to develop close working. relationships with three different agencies in the Government:. of:i Botswana However in spite of the above complications things have improved during the reporting period and many of the problems that were apparent at the beginning, have now been resolved. Particularly important has been the improved communications between the three locations with a telephone being installed in the ATIP... Office in Francistown. Also of great help has been the purchase :of another project vehicle (Peugeot station wagon) for commuting between Gaborone and Francistown. Also -related to communication has been the fact that all the USAID technicians except the Team Leader have had the opportunity to take a short course in SBtswana .or. Kalanga
-although no one as yet has become fluent in either of them. Another; major improvement has been the purchase of two more Apple microcomputers. (Apple IIe's). They have been located at Mahalapye and .Francistown'.. Since their purchase in May the process of analysing data has been greatly speeded up. Looking back it is a pity that perhaps these purchases were not made much earlier since it would have facilitated more timely analysis of surveys' and trials.
Administrative .tasks'still continued to occupy a great deal of Norman's time during the second year of the 'project. It is difficult now to see what further can' be done to .cut. down the proportion -of time spent on administrative matters. As :indicated above the- complications arising from the project being scattered over different locations :and dealing with a number of agencies makes administrative -work inevitably burdensome. However a good deal of information has been gleaned during the reporting period on operational procedures in different agencies. The problem still arises :.that personal intervention by Norman seems to be necessary in resolving many of the problems that arise. This means inevitably a commitment of time -- a scarce resource!
It is. pertinent to raise the issue "that was:also discussed inithe 1982-83 Annual Report about the problem o'of the general administrative burden. placed on the leaders of most.Title.XII projects including ATIP. Most counties in approving project:leaders iuse- as the major evaluative criteria,, the professional expertise of the'potential' project leader. Thus they expect
File:.AAWS48/Admin 1 3.1 Date: 11410184

such individuals to play a leading professional role in the implementation
of such projects, something in practice it is very difficult to do.
Concern about this problem was expressed by Collinson, a consultant early
in the implementation of ATIP. The problem of differing perceptions of the
various parties associated with projects like. ATTP with respect to the administrative/professional roles of this leadership needs consideration
and resolution. One idea that was mentioned in last year's Annual Report that might help such leaders to play more of a professional role would be for AID missions to have some centralised administrative expertise available for the various projects that could, when necessary, play a short term facilitating role in helping to produce documents, accounts, etc., as required by AID. One of the continuing time consuming and frustrating activities of the current leadership of ATIP is understanding AID procedures, implementing such procedures and producing records and accounts documents that satisfy the requirements of AID. Unfortunately, to the outsider at least, it appears difficult to get detailed documentation of such procedures while, perhaps incorrectly, to the uninitiated, they are perceived to change frequently. A facilitator in the AID missions could greatly help in overcoming this problem.
An example of the above is with reference to obtaining GOB reimbursement for expenditures of funds that AID has agreed to provide through the Research and Operations Support Vote In order for GOB to get this reimbursement the Team Leader had to spend- a great deal of time putting together the, necessary documentation required :for this transaction to take place. It would have been a lot easier if someone-from USAID could have helped to produce the required documentation. Unfortunately the capacity within GOB to do this in a timely manner did not appear to materialise and therefore. considerable pressure was placed on the Team Leader to provide this function. Until reimbursement was requested it was not possible to get new funds allocated to the Research and Operations Support Vote by USAID.. A facilitator from USAID could have helped supervise and check the information that was being put together by the Administrative Assistant and people within the GOB. This would have helped reduce the time spent by the Team Leader, undertaking his function which was. allocated at the expense of time devoted to professional activities.... As a result .of the earlier.problems with reference to ATIP purchasing equipment with USAID funds, a line item was placed in the MIAC/USAID contract agreement enabling equipment to be purchased through Kansas State University. This appeared to be a major step forward but recently there do seem to have been some problems with reference to implementing equipment purchases in an efficient manner. This is largely because of the stringent accounting and purchasing procedures that the State of Kansas currently have and which have to be followed by Kansas State University. This continues to be a problem but it is hoped that. with goodwill on both sides it will be possible to implement purchases through Kansas State University.
An effort is being made to use the Apple III microcomputer for some of the routine administrative requirements. For example, leave entitlements, votes, and ATIP library lists are already kept on disks for quick updating. It is also anticipated that inventories will also be placed on disks in the near future. Such streamlining of administrative work where possible can potentially be of great help in relieving administrative burdens,
File:.AAWS48/Admin 3.2 Date: 17/10/84

Assistance was also provided in early March to the Francistown team to
measure and redraft the rough site plans for the T4 building sites in
Mathkngwane, Matobo, and Maropong villages. The hebd for these plans
became critical because the end of the GOB fiscal year was approaching and
funds allocated for building the tondavels could not be used. until site
plans were approved and permission to build granted.
Hobbs made arrangements .to purchase four ridger ploughs and to construct
two flat-bed trailers:: for .implement transport for the Francistown and
Mahalapye agronomists.
Inevitably the ATIP teams at Mahalapye and Francistown do hive
administrative duties to perform. Hopefully over time these will become
less burdensome. Ways are being sought to reduce the time spent on such duties. Hopefully the clerical staff at both places can assume some of the
Obviously Francistown during the first operational year in 1983-84, have
- had a heavy administrative burden in setting up procedures in.the
Francistown area. Also time has been required to set up the necessary
infrastructure, For example considerable time was spent. in the purchasing
and delivery of building materials for the construction of seven dwelling units in Tutume district. Three units were built in Matobo, the village
furthest from Francistown and two in each of the other villages. A two-room unit was constructed in each village for T4.staff that are
expected to: be assigned to the project. Single-room units were built to
house- enumerators. The third unit in Matobo (single-room) was .built either
to house a second enumerator or to serve as a store room for project
equipment. This activity requited a considerable time commitment from
project staff.
The Mid-Term Evaluation Team spent two weeks in July.,Looking at the
projects. ::The five prion team consisted of C. Francis. (Nebraska/Rodale
Press), N. Flora (Kansas State University), D. Galt (FSSP/Florida), H.
Sigwele (GOB) and B. Whittle (Washington).
There= were a number of issues that ATIP brought .to the attention of the Mid-Term Evaluation Team so that they could be specifically addressed in
their report -- if they felt such consideration was justified. It should
be noted that there was in general no official approval obligating the team
to address these issues but reactions to them were considered to be of
potential value to ATIP members themselves. A list of the ..isues .presented can be summarised as follows.
Because of:
(a). The harsh climatic environment -- giving ,rise to little :flexibility
in the farming system, and
File4.AAWS48/Admin 3.4 Date: 17110/84

especially if clerical staff can be taught to work with them.
Major assigned administrative duties for the Deputy Team Leader remained as
in the last report -- overseeing participant training activities and
managing the KSU Revolving Account.
No additional long-term trainees were sent abroad during the report period but the six already on BS and MS courses in the United States continue to make satisfactory progress. Arrangements were made M. Modise, CPO (Post
Harvest) to attend KSU's Grain Storage ana Marketing short course June
11th-July 27th at project expense.
The Revolving Account continues to serve the project and personnel well.
Again it provided advances against expenses for old and new staff. It provided funds for the initial payment for the Apple Ile microcomputers
that were purchased for use by the Francistown and Mahalapye based teams. Money to reimburse the account for expenditure during the second year (August 1983 to June 1984) was received from KSU. An additional refund was provided by a short term trainee whose overseas expenses. were paid ultimately by USAID. Some clarification has been obtained with reference to the use of the fund. For example, it cannot be used for purchasing major items of equipment, since cheques necessary for paying suppliers have to be issued directly by the State of Kansas. Also it has been agreed that at the beginning of each financial year, that is July Ist, the account will be replenished to an equivalent of 615,000, thereby correcting any discrepancies arising from fluctuations in exchange. rates.
Hobbs spent considerable time arranging for the Ham and Johnson consultancies. In the case of the In-Service Training consultancy, Hobbs spent a week in the field with Johnson interviewing DAFS field personnel. Time was also spent with editing of the consultants report.
Additional time was also spent arranging for the Grain Storage Short Course laid on for DAFS Regional CPOs and some BAMS staff.
Other administrative duties included: procuring 1:50000 scale maps and aerial photographs for the selected village' areas in Tutume District
Arrangements for construction of the Francifstown office building continued intermittantly. Permission to construct this five-roomed office block'% nd separate toilet facility was granted by the Land Board and the Franciiiown Town Council. Currently the. project is held up because the estimated cost of construction far exceeds the money' allocated on the basis of estimiaes given in 1983 by the Gaborone: firms. Increased costs resulted from ,unexpected needs in the building -- additional toilet facilities, a new electric. power source (line) for air .condit6nintig, and establishment of access roads and parking areas. There is also the question as to whether construction should go ahead because of uncertainty where the RAO s office compound might be located in the future:.
Assistance was given in arranging for a motor cycle loan and for a reasonable transportation allowance for the Shoshong .T4 (W,, KeipeilYW
File:.AAWS48/Admin 3.3 Date: 17/10/84

(b). Little technology on the shelf that will work every year -- once
again largely a function of the harsh climatic environment and high
variability in the inter and intra annual rainfall
a relatively high degree of emphasis is being placed on:
(a). Descriptive/diagnosteic and design stages work -- in order to
understand when, anid in what form available technologies will
work, and
(b). Increasingly, research managed and implemented (RM, RI) type work -to provide results in drought years, expand the "knowledge horizon"
of farmers and provide tangible results that improve dialogue
possibilities with farmers.
Is this approach justified?
Three currently of particular importance are:
(a). Should ATIP's mandate for farming systems research (FSR) be
expanded to include policy/developmental issues (ATIP call this the farming systems perspective (FSP. -- see Section 1.I ATIP believe
it is necessary. If so how, can it b6 done? .
(b). To date: the sociological/anthropolo ialt input has been lacking.
GOB has now committed itself to providing someone full time for a
period of one year starting.August 1984. What topics urgently need
consideration bearing in: mind that:
(i.) his input has been rge l acking. during the :first two
years of the project, and
(ii). The FSR mandate of ATIP
ATIP team members are stationed at four locations --Mahalapye, Francistown Sebele and Gaborone -- therefore:
(a). Complicating and increasing administrative burdens which are
further exacerbated by having to deal with GOB, USAIDI/S, KSU and
MIAC administrations. .Administrative burdens are heavy for all Title XII projects. Are there any ideas how this burden can be
eased to permit Team and Deputy Team Leaders -- usually selected on
the basis of professional competence -- playing a more productive
professional role.
(b). Complicating the problem of communication and cooperation. Three monthly meetings of all team members are planned, while currently
each team draws up its own reserarch agenda which is discussed at a ~whole team" meeting. Suggestions on how to improve communication
and cooperation would be welo 6mie. Also is it justifiable that the
File:.AAWS48/Admin .5 Date: 17/10/84

Mahalapye and Francistown teams have their own research agendas?
For a number of reasons -- different farming clientele, high professional competency on each team, and lack of definitive
solutions to problems in this harsh climatic environment -- ATIP
think "individuality" of the two teams is justified but what was
the opinion of the Mid-Term Evaluation Team?
One of the aims of ATIP is of course to help -- along with similar projects
-- to institutionalise the farming systems approach to research (FSAR) -or rather, currently FSR -- in Boyswana. Three issues pertinent to ensuring this are:
(a). Training of Batswana. ATIP are concerned about the lack of person
years currently budgeted for long term training -- see letters
between the Team Leader and the USAID Mission Director.. The
response from the Mission Director specifically authorised the
Mid-Term Evaluation Team to look at this issue and make
recommendations. Currently this is the most significant issue as
far as ATIP is concerned.
(b). Resources for FSR. How much emphasis should be placed on FSR in a
country like Botswana with a small GNP -- largely because of its
low population -- and therefore limited resources for research and
extension? What strategies should be considered in order to
maximise the return from these limited resources? ATIP believes that for this and other reasons it is important to link FSR work
with extension as well as research personnel. How can this be done
in a pragmatic manner? Ideas would be appreciated;
(c). The project is currently designed for five years. ATIP believes
this gestation period to be too short -- given the difficulty of
the climate, relatively little technology on the shelf and the lack
of trained manpower -- to institutionalise this approach. The
opinions and recommendations of the Mid-Term Evaluation Team were
sought following discussions with GOQ and USAID personnel. It is
considered important, for planning purposes, to know very soon hat
the length:.of theproject is likely to be.
Three issues that are relevant here are:
(a). Consider a provision that would allow USAID technicians to attend
professional meetings -- it appears possible for Batswana to use.
ATIP short term training funds for such purposes -- to develop
their competence and to provide the potential for meaningful
professional interaction. There are no funds available for this at
the moment.
(b). Suggestions would be welcome of career paths for the USAID.
technicians so that the USA builds up a large pool of developmental expertise. Only three of the seven ATIP personnel are currently in
tenure track positions and two of those will be retiring at the end
File:.AAWS48/Admin 3.6 Date: 17/1i0/84

of their current assignments. This is a very major issue.
(c). In terms of continuity and to prevent expensive "relearning cycles"
it would be desirable to have an overlap between the incoming and
outgoing technicians on the ATIP project. This does not appear to
be possible under the current contract terms. Can some provision
be made for this?
It is understood by ATIP that most of these issues were considered by the Mid-Term Evaluation Team in their report. However ATIP still have not obtained access to the final report since it still has to be approved by the GOB.
File:.AAWS48/Admin 3.7 Date: 17/10184

Three types of training are associated with the project: long and short term formal training and on-the-job training.
Formal training sponsored with ATIP funds is shown in Table 4.1.
Gaesegelwe P. NADF, DAR DA (Swaziland) 1.979 ES (Agronomy) [511 Jan. 1983
Hehie, F. DAO, DAFS DA (Swaziland) 1979 Is (Agronomy) KSt Jan. 1983
Monyatsi, T. ATIP, DPS ESt (Swaziland) 1981 MS (Ag. Econ.) KSl Jan. 1983
Modiakgotla, E. ATIF, DAR US (V. Illinois) 1981 MS (Agronomy) KSU Aug. 1913
Horemedi, G. ATIP, DAFS DA (Swasiland) 1980 ES (Agronomy) KNHSU Aug. 1983
Tjirongo, M. ATIP, DPS DA (Swaziland) 1982 BS (Ag. Econ) ISU Aug. 1983
Siguele,H I NADP US KS (ASTl) FSAR Workshop (1, itub Mar. 1983
Tjirongo, N. 'ATIP DPS DA (Swaziland) 1982 ...FSAR Workshop- l..5xhmb. Mar. 1983
Tale, B. Director DAFS BSc Visit US Institutions Jul.-Aug.'13
Hasikara, 5. ATIP, DAR CA (Botswana) CIMNYT Workshop Malawi May 1984
Monyalane, P.[. Hort.,DAR Vegetable Course USDA Jan.-Feb.'84
Modise, N. DAFS Grain Storage KSU June-July'84
4 Clerical Staff ATIP DAR Word Processing F town May 1984
5 Counterparts ATIF Daisy Statistics F.'town May 1984
2 Staff DAR Daisy Statistics F'town Nay 1984
18 Staff DAFS, BANE, WFP Grain Storage Sebele Sep 1984
The six participants sent overseas for long term training last .year were still away during the current reporting period. Details concerning these
individuals are in last year's Annual Report. In summary, two each were
sent from DAR, DPS, and DAFS. Four are studying in Agronomy and two in
Agricultural Economics, two of the participants are studying for MS degrees and four are reading for BS degrees.
These six participants are making satisfactory progress toward their degrees There was concern that one participant would drop out and return
File: .AAWS48/Train s < > 4. bater 1/10/84

to Botswana. However, he arranged for his wife and one child to travel to the USA and -remain there until he completes his studies. He now appears to have settled into his studies with a better heart. Funds for the family travel and subsistence are provided by the participant, not by GOB or the project.
Another participant got off to a poor start but he has improved his academic performance greatly. Things look much brighter f-rC"him now. A third one, Tjirongo, has been admitted to the- Honours programme at Iowa State University, as a result of his excellent grades.
No additional formal long-term training was initiated during the reporting period. The project is already two man-years ahead of the planned training objective while USAID funds for training are much more restricted than was believed when the project paper was prepared and when the project team members first arrived in Botswana. Also there is a critical shortage of Botswana funds due to the general recession and to the drought. In addition all but two of the current counterparts who are qualified for degree training are very recent diplomates. GOB policy is to keep new staff in post for at least two years before considering them for further academic training.
In last year's Annual Report concern was expressed that the level of funding for long term formal training of Batswana budgeted in the project appeared to be too low. The reasons for these were discussed in last year's Annual Report.
The project design team originally anticipated that two Batswana would be trained for each expatriate staff member assigned to the project. This appeared to be necessary due to the shortage of trained manpower., Thus 16. participant trainees were anticipated and catered for. Ten of these were to be funded by USAID, through the project, and six by GOB. rt is anticipated that the six trainees already in the USA will need 16 man-years, of training (two MS candidates will need four years and four BS candidates will require 12 yea-sY, This leaves only 6 man-years of budgeted training, or enough for two more participants if they are trained to the ES degree level. If the seed component of the project is activated and two seed specialists are sent for their allotted training (one-half year each) there will not be enough training time left for even two more BS degree candidates. GOB funds for training are very scarce and'are not being., released in sufficient volume to permit additional degree training at this time. So unless additional non GOB government funds are found the proposed project training programme will have to be delayed or reduced. This is very serious because trained manpower is in such short supply in Botswana and training is one of the most effective aids that donor agencies can provide.
However discussions with the USAID Mission Director resulted in an agreement to bring the issue of long' term training to the attention of the,. Mid-Term Evaluation Team. Apparently they-agreed that more funds should be allocated for long-term training. It is therefore anticipated that an approach will be made to GOB suggesting a :modification in terms of USAID; : resources being allocated for long term training.
File:L .AAWS48/Train 4.2 Date: i6/10(84

During the last year a total of 31 people have received short term training. with funds provided under the ATIP project,. F K. Monyamane from DR',went on a six week USDA course. in the United States involving Vegetable Production and Marketing. M. Modise from DAPS, the newly appointed Post Harvest Crop Production Officer went to the Grain Storage course at Kansas State University.
Masikara who is an agronomist in the ATIP project went to Malawi with Siebert to attend the CIMMYT workshop on issues on. Farm Experimentation. The arrangements with reference to F:lsikara are of particular interest and perhaps form a model for the future. As USAID technicians receive invitations to attend professional meetings, it would seem on occasion, appropriate that counterparts could also benefit from attending such, meetings. It was thought that since Masikara could both benefit from, and contribute to, the deliberations of the meeting, that this would be a good use of short-term training funds. It is planned that such arrangements will be continued in the future.....
Ten Individiuals received training on Apple IHe microcomputers. Two courses each of two week duration were held by a short-term. consultant, Mrs. Sandra Miller, in Francistown for ATIP personnel. Two Apple,Ie domp'uters were.-.. borrowed from the Apal .Agency in Gaborone permitting Mrs:., Mifler to instruct a neither of people at the same time. The first -durseinvolved instruction in the operation of the microcomputers plus the use of the word processing package utilised by ATIF personnel. This is Applewriter. The typists in Francistown and Mahalapye plus the administrative assistant and trainee at Sebele attended this course. The second two week course involved once again the use of the Apple Ile microcomputers plus instructions on operating a statistical package called Daisy Professional. All the counterparts, with the exception of Masikara, who was in Malawi, plus the trainee in Sebele, another person attached to the Cereal Breeder at Sebele, and one of the USAID technicians, were involved in this course. There is no question that these courses have been extremely valuable in helping staff to overcome fears concerning using microcomputers and unquestionably will result in increasing the productivity of the microcomputers. Mrs. Sandra Miller prepared extensive teaching materials which are available for future ATIP personnel to use.
Another in-service training course was held in cooperation with the Food and Feed Grain Institute at Kansas State University, which was attended by 18 individuals from DAFS, BAMB and the World Food Programme Considerable amounts of grain are known to be lost during on-farm storage in Botswana. These losses become increasingly serious as demands for food increase and as the production of grain lags behind needs. Accordingly there was a request for an extension programme on improved grain storage techniques. This led to the development of a plan for an in-service-training course in on-farm grain storage (see Section 6.2.2 (d)).
Assistance was also provided by Hobbs and Ramolemana in some of the in-service training courses planned by the Regional ALDEP Managers and CPO's at the Regional Rural Training Centres. This assistance will continue on into the next reporting period.
File: .AAWS48/Train 4.3 Date: 1610/84

This is an area of great importance since FSAR training is not an established part of the degree programs. In the project design stage it was agreed that when counterparts proceeded overseas on long-term formal training programs replacement counterparts would be provided so that on-the-job training could continue, while at the same time the productivity of USAID technicians could be maintained. In general t : -'Gdernment of Botswana has been very good about providing counterparts. in addition during the JanuarylFebruary break a total of four diploma level BAC students were assigned to. the roect.
In general the approach to .on-the-job training is to assign tasks that provide professional experience including conducting structured and unstructured interviews., survey supervision, direct field measurements, collection of technical data at the field level, and data analysis. In addition on occasion on-the-job training has been supplemented by discussions involving the concept methodology of FSAR and the assignment of reading materials in specific disciplinary areas and FSAR.
In addition to counterpart on-the-iob training provided by the expatriate project staff, plans by Hobbs and Ramoler.ena are progressing to assist the Department of Agricultural Field Services with some of its in-service-training needs.
File: .AAW4$/ Ttain 44 fa.te: 1610/84

Ag. Economist: D. W. Norman
Most of Norman's time spent on professional work has been devoted to supporting other team members.
The research activities have been somewhat limited in scale and have simply involved helping professional activities of other profet members. Little field work has been done by Norman this year. Some help has been given to designing survey forms. During the cropping season an effort was made to travel both to Francistown and Mahaiapye at least once every two weeks. These trips served to help provide a professional input while at the same time dealing with administrative matters. An effort was made to continue these visits at two week intervals during the dry season when data anaylysis receives priority:.
The major research activity continued to involve working with the Apple III microcomputer. As reported last year a number of software packages were purchased that are compatible with each other and allow Interfacing across the SOS operating system on the Apple III to the DOS system on the Apple lie's and the Apple III, when operating in emulation mode. The equipment and the software packages that have been purchased for the new Apple Iles and Apple III are shown in Table 5.1 All the packages interface via the DIP system. ATIP still does not have a good statistical package for the Apple III. However this is no longer an important issue since it is anticipated that most of the data analysis will now be done on the Apple File's in Mahalapye and Francistown. With this in mind two statistical packages have recently been purchased. These are Daisy Professional which is an update of the earlier Statistics with Daisy package and Statpro. There are still a few bugs to work 'out with reference to the Daisy Professional package but it is already proving to be of considerable value. The Statpro package also .looks to be promising although as yet it has not been possible to find a way for it to read a DIF file This is contrary to the information that was obtained concerning its capabilities before purchasing the package. It is anticipated that this issue will be resolved in the near future, possibly with the help of a consultant writing an interface programme.
In summary Norman's work with the microcomputer has involved the following:
(a). Continuing to learn' how to operate the software packages purchased
for the Apple machines.
File: .AAW S48/Seb -5. i Date: 17/10/1984

(b). Continuing responsibility for supervising entry of the data from
surveys and trials on the Apple III machine at Sebele after which
it is downloaded into files suitable for analysis on the Apple
lie's machines at Mahalapye and Francistown.
(c). Continuing to provide training to the microcomputer operators at
Size 128K 128K
Monitor 12" Screen Kaga Denshi 12"
Drives 2 Drives 1 Built-In Drive
I External Disc Drive
Hard Disc Pr e1 Profile (5 Megabyte)
Pointer 13Column NE Printer lun 132 Column PrIsm Printer
Constant Power Supply Voltage Regulator Omnipower. 1152
TYPE OF PACKAGE --------- ----------------APPLE II APPLE III
Data Base Management System Visifile Omnis
Calculation (Spread Sheet- Visi-alc Advanced Visicalc
Graphics VisipIot./ Visitrend Business Graphics
Statisti cs SPS
Daisy.. Professional
Stat. pro
Word Processing Applewriter 11I
Note: The list only includes packages used for handling data Supportive
packages such as Catalyst, Backup III, Zoom Graphics, etc are not
included in the list. All the packages can inte-rface with each other
via the DIF system and it is possible to cross operating systems -from SOS on Apple III to DOS on Apple I and vica versa.
(d). When possible helping other staff members of DAR who required data
entry and analysis. This work has involved three staff members:
(i). Entering data from fertiliser trials carried out by the
Soil Scientist DAR (Stewart Jones).
(ii). Entering results of a vegetable consumer demand survey
undertaken by the Horticultural Economist at DAR
File: .AAWS48/Seb -5.2- Date: 17/10/1984

(iii) EnteLin(o data an suraohum varietal testing that has been
cul.7 cted over the ears at Sebele and is been analyIsed by
the~ 17IIORM1L agrono~mists (Carter ?nd '?oungqulsti.
In general the microcon uters ars now working satisfactorily although, as indicated, there -re a few prollbLims -with some of the softwar -e packag-,es arid also continuing problems with thie printers Mt Mahiaaove and F'rancistown.
There does seem to be a case ft- naint a cosuiat to work out some -; these jeinor problems. Negotiations have been. going on with Kansas State University for several montlis to try Anid ttu once more as a short-term consultant. Oricnallyv it was intended that lie would do this work while at Kansias Stat~e Univet-sity, "jut there appear to have been some hitches with reference to Makillg this arrangement. Thie:efore consideration may have to lic made to get Ihin to visit Botswant -again in ucder to sort -,ut remaining relatively minor problems.
In addition to a ntambur of meetings in Botswa,-na, ~tlhrman als4o mInde two trips overseas during the reporting period. In Decemeber he participated in an in-house workshop on the futtu direction of On-Farm rlese~rch in the CIAT Bean Programme at CIAT in C;41i, Colombia, and also attended as member a meeting of the Comite Sciuntifique tt Technique, Irnstitut Senegalais de Recherche Agricoles, in Dakar, Senegal. In May hie attended an Acriculture Development Council sponsored meeting at Beflagic in Italy on Intra-Household Piocesses and Farinn Systems Aniilysis. At this confere-ace a paper written in conjunction with Baker was given. Also in May lhe spent a short period in Tunisia at the request of the MIA: or o~ect that is currently underway there helping thera on plans for farming systems project. While there seminar was also given.
The proposed work pla-n uf the Team Leader is as follows:
(a).To cntine ,jot responisibility~ for the administration of ATIP
(b) To coiatinue working with "he Apple 11-3 microcomputer in order to
establish effiutient opera;ting procedures and to facilitate
timeliness with reference to Lhe data collection processing
a:.nalysis link.
(c) To play greater rule in aiding the professional Work of the multidiscipliftiar Dlalalapye and Fraicistown groups and help meld
them into: truly intetdisuipJlinaty farming system teams.
(d) To continue encouraging increased cooperation between the different farming systems pjetand bt.,t.en emaperimernt station based
scientists includino the 1.14SORM'IL and Bean/Cowpea CRSP
scientists ;, nd farming system project, scientists: and to seek
ways to improve linkages with rpolicyidevec*oment agencies in order
that the farming system perspective -- that: is, policy issues
can be iricoruorated into the work. of the iaruting systems teams.
File AAW;;48/Sub ~ -0. 3 -Date: 17Il1011984

USAID. J. A. Hobbs (Agronomist)'
GOB: G. Ramolemana (DAFS, September 1983- Current)
The duties of the RELO/Agronomist remain essentially extension oriented but some assistance was given in setting up research activities in the SFrancistown and Centrail Regions..
Hobbs and Ramolemana assisted the new Francistown team members as they selected villages for field activities. Later .they accompanied the agronomists on a field inspection trip with the FAO Soil Scientist to learn the variety of soils that will.confront them in. Tutume District. Later Hobbs demonstrated the use of the soil. penetromter c.:at both Francistown and Mahalapye and discussed with the Francistown staff how to select sampling sites in experimental plots and how to assess; crop perform since in plot quadratst.
Th4 decision to expand liaison activities beyond the Farming Systems project areas has proved not only to be wise but is also extremely ,challenging. The continuing drought haIss reduced devilbpiefit of FSR research -resuits that can be .extended to DAPS' staffassiditk with the projects ahd clearly much needsto be done to improve .c.tii nications :between extehsion and research over the whole of BotswanArit
(a).- The results of the survey. made by visiting RegionaI anid District
Agricultural officers.about problems facing extension personnel and farmers in their efforts to improve crop. production were developed
into a questionnaire which was distributed widely in the
Departments of Agricultural Field Services and :A griuittUral
Research (Arable Research Division) and to som non-ministry
people. Return of questionnaires was slow, but data" from them was
put intd the microcomputer in August and is currently being
analysed. It is expected that such an analysis will indicate what
are considered the most serious problems and those needing most
.:-immediate attention.
(b). Visits to District Monthly Management. Meetings were continued. At
each visit an attempt was made to point out fatets' of liaison
activity being used or to talk about information coming from
research that may have application on th farms of the' district
File:.AAWS48/Relo 6 1 Date i7/10/1984

At'each meeting an offer was made to go to individual extension
areas on invitation, but few ADs have so far asked us to visit
(c). Meetings with the Regional Crop Production Officers at their
headquarters were started. The first meeting was with the Gaborone
CPO and Regional ALDEP Manager. During this discussion these
officers indicated that research officers sometimes have different
ideas and philosophy about certain agricultural practices from
those being promoted by the extension people. This caused them
concern. They asked if it would be possible to meet with research
officers in crop production areas to get squared away so that both
groups are telling the same story. A meeting was arranged at which
the Chief Arable Research Officer and representative crops research
officers met with the Chief Crop Production Officer, two crops
extension officers, some other agricultural extension officers, an
AD from Gaborone Region and representatives from the Agricultural
Information Section.
As a result of the dialogue several questions were raised.
(i). Are Agrifacts and other forms of printed extension
information considered valuable and are they read and used?
(ii). Do the current Agrifacts need updating and correction?
(iii). Are there areas for which new Agrifacts should be prepared?
A decision was:made to have the CCPO :as k all Regional i CPOs .- and
all ALDEP Regional Managers -- to review current Agrifacts for
content, accuracy, and readibility and to have each-officer solicit
opinions from a few ADs. in his/her Regiaon at this"iater. CPOs
were asked to send their critiques to the CCPO by the end of March.
These would then bet forwarded to the Chief Arable Research Officer
.(CARO) for attention. of research staff,, autihors of most of
Agrifacts. A later ;meeting of .CPOs, CROs, and Information Officers :was suggested for the' end of Juhne to review comments and e sponses
by research and to plan necessary corrections, improvements, and updating of current Agrifacts and to plan new ones. This was an
agenda item on the programme of the RCPOs conference (see Section
6.1,2.1(e) below).
In addition a survey was undertaken to determine how many ADs and
other extension officers did not have access to a set of Agrifacts
(Extension Handbook), Whereas everyone in extension administration assumed that each AD had the Handbook, in actual fact only a very
small proporation had it, and many did not have a single: Agrifact
that they could lay hands on. This needs to be remedied and steps
are being taken to do so.
(d). Visits to research projects and stations continued during the
report period. Ramolemana visited IFPP and attended the MDP and
NADP FSR Workshop at Maun/Gumare 6th to 11th February. Hobbs and
Ramolemana visited Motopi and Mahalapye research fields and the
Dryland and EFSAIP activities at Sebele.
File:.AAWS48/Relo 6.2 Date: 17/10/1984

(e). From contacts made. so far and from discussion with many extension officers it appears that officers in DAFS get little counsel and
guidance in the organisation and performance of their duties.
Several have indicated that essentially they have been given a copy
of their job description and told to get on with the work This
has been as true of more senior officers who come to. the Ministry
with no extension experience as with junior men who come with the
Agricultural Certificate and a little experience they picked up
when assigned to an AD in the field during their certificate
training.. As a result most officers flounder around before they get their feet firmly on the ground, some take much too long to
become really effectivei and a few never really make it. This is unfortunate fo two major reasons. First the presentation of the
extension message, which -is so desperately needed by the farm
families of Botswana, is drastically delayed. Second the lack of knowledge and effectiveness inevitably affects the morale of the
officers so that they become disinterested in, or even:prejudiced
against, learning their duties. Some potentially useful officers
may become permanently disallusioned and disgruntled, and may fail
to become useftU 'workers. For example the crops officers in the
-regions and 'headquarters are assigned the major responsibility
for extension-research liaison in the present departmental setup,
but With insufficient knowledge of their duties and of research
officers activities 'and the technology that develops from them they
are not able to effect communication between these groups.
In order to -develop a better understanding of the activities of a crops officer and to engender a higher esprit de corps Inthe
staff a plan was developed to have semi-annual conferences and
workshops for the regional CPOs and HOs and for those in the' Crop
Producion Divisibn headquarters' staff. At these meetings
discussions on responsibilities, isteful activities, extension
methodology, and technical information will be organised, at times
led by personnel from the field or from headquarters, some,1Ums
using resource people from administration, research, or other
organisations .or agencies. The first of these conferences was held
in Gaborone. 25th-26th June with, 26 crop officers and 17. visitors
present Hobbs acted as secretary and has prepared a summary of
the conference.. IN COUNTRY TRAINING
(a). The report on the Farming Systems Workshop, i held in 1983 was
completed during the reporting period. It has been' printed and
(b). In-service-training courses originally given by ALDEP personnel to
familiarise ADs with the ALDEP packages. and with cultivation
equipment assembly anrid maintenance were expanded during 1983 with the appointment of the ln-Service Training Coordinating Committee,
of which Hobbs is a member. The ALDEP Regional Managers were on
post for tthe first time in late 1983. Responsibility for these
broadened courses and workshops was assigned to them and their
counterparts, the Regional CPOs. Under the initial guidance of the
File:.AAWS48/Relo 6.3 Date: 17110/1984

Coordinating Committee and with input from both ALDEP Managers and
CPOs a-broader base :of instruction has been developed. Project
officers including the RELOs have been asked to assist with these
courses. During this reporting period, however, only a few courses
were attended because of short notice and absence during course
dates. Hobbs met with a group of Central Region ADs at MRTC in
September when he discussed fertiliser recommendations.
Observation of the courses set up for the Gaborone, Southern, and
Francistown Regions in 1984 shows that nearly all of the
instruction, apart from that relating, to ALDEP programmes and
packages, is provided by non-extension staff. Senior extension
officers are not taking a hand in instruction at these courses,
they only assist by organising and administrating them. Much of
the instruction has been assigned to research officers from Sebele.
This did not pose too serious a problem in the Gaborone Region, but
an officer in a popular research area like entomology or soil
fertility will find it difficult to take time off from his research activities to give a discussion of his subject matter area in all regions or even for single courses held in Maun or Francistown
where long distance travel and considerable travel time are
It seems obvious that the time the crops research officers have
available for "training" should not be spent teaching- the ADs directly, but rather should be spent teaching Crop Production
Officers, who will then be in a position to transmit this
information to the ADs at formal or informal training sessions.
This emphasises an obvious fact -- that all levels of extension
workers need additional training if they are to do their job well
and if the extension activity is to be truly effective and
(c) The need for in-service training in Agricultural Field Services is
supported by nearly every staff member in the department, but the
present training effort is limited in scope and is very sporadic.
Consequently a suggestion that ATIP should sponsor an in-service
training consultant was well received. Arrangements were made for Dr. Robert L. Johnson, Assistant Director (Personnel), Cooperative
Extension Service at Kansas State University to spend four weeks in
Botswana 6th June to 3rd July. He spent about two weeks visiting
S all levels of extension workers at headquarters, in every Region
S and in: many Districts of the country. 'The remainder of his time
was spent developing a suggested programme of in-service training
for the department. His report was submitted and has been printed.
Copies have been sent to Divisions', and Regional and District
-;'offices in DAFS.'
(d). The need to control losses of grain while in storage on farms has
.. been emphasis d by the. distressingly reduced yields of grains being
produced aiid harvested in the current drought :ears and by the
constant ai r apid increase in demand for food resulting from
increases in population. The recognition of this need resulted in
the appointment of a Crop Production Officer (Post Harvest) at
Headquarters in .198q and in the development of a plan to provide
File:.AAWS48/Relo 6.4 Date: 1711011984

ADs with the best knowledge available about on-farm grain storage.
A project, proposed and strongly supported by ATIP, was developed
to provide special training for the Post Harvest Officer and later
to offer in-service-training on On-Farm Grain Storage for Regional Crop Production Officers. The Post Harvest Officer's training was accomplished by his attendance at Kansas State University/USAID's Food and Feed Grain institute's short course on Grain Storage and Marketing, 11th June to 27th July, with course expenses paid from ATIP short term training funds. The course for regional CPOs was
held at BAC 3rd to 14th September. Instructors for this course
were two specialists from the Food and Feed Grain Institute -- Ms.
R. Burroughs and Dr. V. Wright -- and the Post Harvest Officer.
Financial support for the outside specialists was provided by the
Food and Feed Grain institute funds (USAID).-. Eight warehouse men
from BAMB also attended this course, together with some individuals
from the World Food Programme. FSR ACTIVITY COORDINATION
Two types of Farming System (FS) Project meetings were organised or attended during the report period.
(a). The second semi-annual FS project coordination meeting was
cancelled, as such: because of the special week-long workshop
organised by the Molapo Development Project (MDP), the Ngamiland
Agricultural Development Project (NADP) and the First Community
Development, Project at. Maun and Gumare 6th to lith February. This
workshop was well attended by staff from- the more southern'
projects. ATIP was represented by' five staff members from
Sebele/Gaborone, 'Mahalapye, and Francistown. Ramolemana attended
and represented the 'Research Extension Coordinating Unit (RECU).
All staff, gained i useful insight into the research and extension activities generated by the two northern FS Projects and by the
Firs Community Development Project. .
The second Botswana FS Project: coordination meeting was held in
connection with the Arable Crop Research Review Conference lith to 13th, september. This meeting was held 14th September, immediately following, the Review Conference. The major issue discussed was the
institutionalisation of farming systems work in Botswana. It wak
decided that the Directors and Permanent Secretary in the Ministry
of Agriculture would be approached with a proposal that a definitive paper on the instiutionalisation issue should be
(b). Hobbs represented ATIP at two coordinating meetings of the British
sponsored FS Projects and the Dryland Farming Research Scheme
(DLFRS) held 21st October, led by DLFRS, and 17th February., .led by EFSAIP. These two meetings were to be, the last' to be led by these two projects as they are winding down and will terminate shortly' after field work ceases this season. Actual termination date isr' i .
1985. These two meetings were forums for discussing plans for..
project completion writeups and staff dispositions. OTHER COMMITTEE ACTIVITY
File:.AAWS48/Relo 6.5 Date: 17/10/1984

Hobbs did not attend the meeting of the Arable Research Priorities Committee that was held 15th August or that of the Arable Agriculture Development Committee held August. These meetings were called without much notice and he was on leave at the time.
The In-Service-Training Coordinating Committee has not.met since October, 1983. VISITS
Hobbs visited the Pennsylvania State University/USAID Farming Systems Project in Swaziland in December and the Washington State University/USAID Farming Systems Research FProject in Lesotho in April.
The .:major problems interferring with the development of. a more effective and efficient extension effort in Botswana are still low level of communication between extension and research workers, lack of technical and extension methodological information on the part, of ,the extension workers at all levels, and low morale of the extension staff. COMMUNICATION BETWEEN EXTENSION AND RESEARCH
In the present administrative setup in the Department of Agricultural Field Services the major instrument' 'to establish and improve .cotitunications between extension and research in cropproduqtion, is supposed to be the specialist crops officer (CPO, HO TO (Hort), etc) in headquarters and at the regional level. Either these officers have not been alerted sufficiently to this facet of their responsibilities, they:feel inadequate or unwilling to undertake the :tasksy, or they believe research does not have information that can be useful to them and is not capable of tackling their problems. Whatever the cause, the Chief Arable Research .'Officer, DAR, states that he receives at most two or three requests per year for answers to farmer problems and/or suggestions of problems needing solution from crops officers or other extension personnel. It is true that ADs seldom ask for answers to their problems and they do not often pose questions needing solutions to CPOs. But it is likely that this situation can be remedied if the CPOs begin actively to interact with ADs and even farmers in these matters.
It seems obvious that inserting another layer of administration (in the form of a RECU) by itself is not likely to improve this communication greatly or rapidly. Rather the requirement that the specialist crops officer serves as a link beween research officers on the one hand and village level extension workers and farmers on the other needs to be highlighted and emphasised. One way to do this is to organise meetings and seminars for crops officers at which this aspect of their work is dealt with as well as discussing and developing techniques and ways that liaison may be improved. A first step toward this type of activity was taken in June when the first extension crops' officers seminar/workshop was held. It is hoped that these get-togethers will be fruitful and that semi-annual conferences, seminars, or workshops will become a regular feature of the annual extension programme. Te second seminar is planned for 3rd 4th
File:.AAWS48/Relo 6.6 Date: 17/10/1984

In the first annual report of the project mention was made of the need for
improving and upgrading the fund of knowledge of all extension workers. it
has become increasingly obvious that workers need extension methods
instruction as well as information on technical matters.
There is great need for more academic preparation for all levels of
workers. Funds for long term training within ATIP are severely limited, but attempts are being made to locate additional training funds. Even if
additional funds for this training can be found there will still be need
for regular update and renewal instruction, and in the absence of long term
training funds this in-service-training activity becomes just that much
more important. This is necessary so that each worker can develop his
expertise and his confidence in order to do his job. better.
This is the background which prompted the appointment of a consultant on
in-service-training for the Department of Agricultural Field Services in June. Project personnel hope that this exercise provided an effective
impetus to the training programme being developed by the recently appointed
Training Coordinator in the Department of Agricultural Field Services. LOW EXTENSION WORKER MORALE
Many, if not most, extension workers in Botswana feel they are ill equipped academically and by orientation and experience for their posts. As a result of this, and. because they often feel isolated from their supervisors and administrators, they have a low opinion of their ability to develop meaningful projects and programmes for their areas, districts and regions. The fact that government housing 2 for ADs particularly, is poor or even not available and that equipment, transport, and financial support is minimal adds to their insecurity and causes real problems.
Even after a good in-service- training programme is started, and an expanded scheme of long-term.. training is organised, low morale.may still be a serious problem among extension workers unless the individual worker is made to feel that his/her work is considered to be. really important by his superiors and by government in general. Real enthusiasum for the work and effective activity can only deviop when the workers are confident of their prowess and when they are sure they are doing an important task Improvement in crop production will be minimal as long as extension workers are unsure of themselves and feel they are unsupported in their work. Remedies for this situation must come from DAFS administrators. Administrators and supervisors also Must explain clearly and carefully to every new staff member what his/her new responsibilities. are.. They must provide proper orientation for the pbst, and give assistance and counsel regularly during the first :.few months on the Job. This is seriously lacking now. Also the administrators must evaluate fairly and equitably the work of each staff member, reprimanding when necessary, praising where possible, and rewarding each as this is possible in line with the quality of his/her performance... Supervisors and administrators..must also make a concerted and consistent effort to visit staff in the field to show their interest and support. Administrative and technical problems raised
File:.AAWS48/Relo 6.7 Date: 17/10/1984

regularly by field men need to be discussed and plans formulated to tackle those that can be solved.
The proposed work plan of the RELO/Deputy Team Leader and his counterpart is as follows:
(a). Participate in administrating ATIP: including overseeing the
project's Revolving Account, supervising the participant training
programme, and consulting regularly with the project's agronomists
concerning their field activities and assisting them in adapting their research results into useful and acceptible farm practices.
(b). Visit District Monthly Management Meetings regularly to keep in
touch with farmer/AD crop production problems and needs and to
inform ADs of new developments in crops research and in extension
(e). Continue monthly visits with regional crops officer teams (CPO, TO
(Hort), ALDEP Managers) in order to help. them obtain required
research information and develop improved crop production practices
for their regions. At times this will involve joint visits to
research stations or fields, and/or visits to FSR field locations.
(d). Continue t'6 develop liaison between crops research and extension
officers and agricultural information specialists in order to
improve current extension publications and to develop needed new
ones. In this regard an E:xtension Publication Requirements
Committee should be organised Responsibilities of this committee
would be to ascertain new publication needs and- to supervise
development of these.
(e). Continue to spearhead liaison between the: country's farming systems
projects in order to coordinate activities. and ensure that all resekirch results are readily available to all projects and that
information about acceptable new practices are-. rculated to all.
(f). PartiCipate in the deliberations and activities f the Arable
Agricultural Development Committee and the Ar.bieResearch
Priorities Committee.
(g). Work actively with DAFS's .In-Service-Training -Coordinating
Committee and with its Training GCoordinator in order to help
develop appropriate in-service .-and other training programmes for
- all Departmental personnel:.
File:.AAWS48/Relo 6.8 Date: 17/1011984

USAID: D. Baker (Agricultural Economist) J. Siebert (Agronomist)
GOB: C. Tibone (Miss) (DPS, September 1983 Current) J. Luzani (DAR, January 1984 Current)
J. Lesothlo (DPS, August 1984 Current) [Works also in Francistown3
The primary research activities planned for the 1983-84 season by FSAR Stage were:
(a. Description and DiagnQostic Stage;
Whole farm:studies
"New village" description-diagnosis
Cowpea baseline study
Regional perspective study
(b). Design Stage (RM,RI) five trials
(c). Testing Stage (RM,FI a.nd FM,FI): six on-farm trials
In addition, it was anticipated that some time would be allocated to preparing reports based on the first year's research.
Progress made by activity is described briefly below. Several unplanned research activities were carried out in response to emergence of new issues and unforseen circumstances (particularly a second year of drought, during the planting season). These activities are also described below.
Four distinct, but inter-related components to whole farm studies were planned:
(a). A multiple visit.. resource use survey.
(b). Technical plot monitoring for whole fields.
(c). Whole field level monitoring.
(d). Special plot technical monitoring.
The first two components were carried out together on 27 farms. In light of drought conditions which prevailed during the critical planting period
-- November through January-- 12 of the 27 farming households participating in components one and two of the whole farm studies either did not plant or abandoned the few plots planted in advance of weeding. It
File: .AAWS49/Mah., 7.1. Date: 15/10184

was decided, therefore, to continue technical ploct monitoring, rather than shift to whole field level mo-itoring, on those "trials" farmers from year one who planted in year two. Because the extra time spent on technical plot monitoring and, more importantly, a substantial increase in the number of plot observations, it was thought necessary to carry out special plot monitoring on fields of farmers who had not yet cooperated with ATIP. The advantages of special plot monitoring, as outlined in the annual work plan, would in any case have been limited due to the abnormal pattern of rainfall.
In addition to planned Whole Farm Study activities, three inventory questionnaires were administered to farmers participating in the MVRU survey.
(1). Multiple Visit Resource .Us1e Survey
The Multiple Visit Resource Use Survey (MVRU) was administered to 17 farmers in Shoshong East and 10 farmers in Makwate. During the first two months of the reporting period, a structured, close-ended, pre-coded survey instrument was developed. To facilitate use of the survey instrument, a detailed instruction booklet was written. During October and early November, the three ATIP Mahalapye field enumerators received training in use of the survey instrument.
Several minor problems were encountered in the survey instrument during the first few months of the season. in conjunction with A'TTP Francistown, the survey instrument was modified. during November and December in order to reduce the time required for an interview and to eliminate coding distinctions which were difficult.. for. farmersm:and enumerators to understand.
The survey instrument in use since December is comprised of a inaster questionnaire and 14 follow-up sheets. Each follow-up sheet _is tised to .record required flow .data:.corresponding to an affirmative response t an activity question on the master questionnaire. For eXamis i 'i a person said "yes" they brewed beer affirmative response to question seven on the master questionnaire --- then the enumerators ask the follow-up questions on Sheet 7. Each f owAI.-up calls for five to ten data entries. The follow--up sheets are as follows:
(a). Fieldwork by Household Members
(b). Fieldwork by Non-Household. Members
(c). Draught Power Used
(d). Non-Labour Inputs on Household Fields
(e). Field Off-Take and Threshing
(f). Fieldwork on Non-Household Fields
(g). Non-Cropping Activities
(h). Use of Household Animals
(i). Sales
(j). Purchases
(k). Miscellaneous Revenues
(1). Miscellaneous Expenditures
(m). Livestock inventory Changes
(n). Household Maintenance Activities
File: .AAWS49/Mah.1 7.Z Date: 15110/84

The average length of an interview is 15-20 minutes, not including a field visit if required to collect technical plot monitoring data. Each enumerator interviews eight to ten farmers twice weekly.
Copies of the master questionnaire and follow-up sheets are collected and replaced each month. In addition, each enumerator fills-in a "problems and notes" sheet for each farmer, indicating any situations that arise which do not easily fit into the survey format. Finally, field maps and household census listings are kept by each enumerator to enable verification of labour use by household members versus non-members and allocation of crop production inputs to proper plots (where feasible).
Despite some supervisory and personnel problems, an acceptably complete and accurate data set has been. obtained from most of the farmers participating in the survey. The data set should enable proposed cash flow, income, and resource use analyses to be carried out. It is questionable whether sufficient detail has been collected to allow plot by plot analysis, but plot specific dith were collected on complementary.survey forms as part of the tehnical plot-moritoring, component of. the whole farm studies
Data from the 1983-84 season are now being entered on the Sebele computer. In advance of analysis of the compete data set, some data have been entered and-analysed at ATIP Mahalapye,. Freliminary analysis has been carried out to accomplish two objectives:
(a). To develop a preliminary profile of farmer activity and cropping
outcomes for the MVRU farmers, to pthent at the Sebele Arable
Research Meetings (held in September, 19S4).
(b). To evaluate cashflow and labour use patterns in order to better
judge the need for continued semi.-weelCiy resource use monitoring.
Preliminary results on days of ploughing, area ploughed, and sorghum harvested in the 1983-84 seaon for the 27 MVRU farmers are presented in Table 7.1. Comparative results are inc uded for Shoshong versus Makwate, male versus female headed households, lih seholds with more than 35 cattle versus those with fewer or no cattle, and for households controlling draught resources (own or borrow) versus-Those which were dependent on other households (hire or obtain through cooperative arrangements).
The table confirms several general impressions: gain: through informal interactions with farmers. 'lakwate farmers, male headed households and draught "owners" ploughed many more days on average than did Shoshong households, female households and draught dependent households, respectively. Sex of head and draught access were strongly, but not completely related, and both factors seemed to :nfluence days of ploughing. Many female households lacked the labour to plough frequently even when they had draught power. Cattle assets did not lead to more days of ploughing, largely because several richer households in Shoshong have turned to tractor hire as a preferred form of draught access.
Even though Makwate households ploughed many more days, they ploughed only a hectare more per farm. Donkey farmers, dominant in Makwate, ploughed less area per day than did the cattle and tractor farmers in Shoshong. Male headed households, asset rich households, and draught "owners" all
File: .AAWS49?/Mah1 7.3 Date: 15/10/84

ploughed more areas than their comparison groups:. in total, per person ploughing, and per consumer. equivalent. The sole exception was in terms of area per day. Draught dependent households pbLoughed a larger area per day
than did owners, reflcting large areas per day ploughed by tractor hirers in Shoshong.,
Most households in both villages ploughed at sqome point ibut'only half harvested any' sorghum. Aaain, male headed households-, rich households, and
draught owners did much better than their co-anterpairts, in terms of the proportion of households harvesting, total amount harvested, amount harvested per hectare, per consumer equivalent,. and pear worker equivalent. The :difference was particularly dramatic depending on draugghit: access. For
all households cambined,,_apprcximately 336 kgs of sorghum were produced in
total. This was more on..average than the !982-83 s ea so n but yields per planted hectare of 74 L-gs reflected the drought co nditions which prevailed during much of the season.
VARIABLE -- - ---- -- - - - - -- - - -
%FARMS WHICH PLOUGHED 82 13C 14 80' 1.06 $1 100 734 78
DAYS Of PLOUGHING 4.1 10.8 8.2 3.7 7.1 .6.2 9.8 1.9 6.6
Toa 3. 4 '4.'3 4.S 2.5 5.7 2.4 5.2 t.? .3.8
Per Person Ploughing 1.4 1.1 1.5 0.9 1.8 0.1 1.5 0.6 1.3
Per Consumer Equiv. a0.6 0. 6 0.7 G.4 0.9 0.4 e.g. 0.2 0.6
Per Day. 1.2 0. 4 0.9 0.8 1.3 0.5 6.A 1.0 0.9
HARVESTING AX! SORGHUM- 5 53 3 '40 7 31 6 2? 48
Total 287 420 404 131 488 232' 532 51 336
Per Coser equv 64. 87 89 42 72 75 98 25 74
PrCuueEqi. 44 55 62 26, 74. 4 74 5 41
Per Worker iquiv. 117 141 '16 1 44 .5? 618 175 I'l 130
a Kgs. sitghun based on 'a s .tandard c' Arsic 1n of 70 kqs. per bag" o f Ith reshed gr ain.
:t~hlsuer units weights: ag '.0 g .0ge ea01 .5 adult woman 0.90i
adult men = 1.00 These weights are based on FAG estimatted caloric requirements.
Worker euivalents derived as, fa lows: each person active in' ploughtplanting 113i each 4titive in weeding 113i each &etive in harvesting 113.,
Table 7.2 presents an overview of average monthly ,zash. inflows and outflows for the 27 MVRU farmers during 'the period Novemnber 1983: to June, 1i984 -the cropping' season. The o ni major cashflow,, not included is traction hire
File:' .XAWS47/Kah. 74-'Dt: L~O8

VARIABLE --------- -------- -------S0890K:N YAKWATE KALE FEMALE -.11 0-3S ALL
Crops 1.4-1 0.07 0.46 1.14 0.21 0.70 0.70
Livestoct 53.2y 12388 G7.15 8.09 81,11 10.10 3?.46
Beer 27.68 .0 9. 44 34.54 2U.3 13.94 18.49
Other 2.06 0.21 1.78 0.83 2 8? 0.42 1.44
SUB-TOTAL 84.07 15.74 68.83 44.0 111.0? 24.16 IC.0?
Gifts and Loans 19.65 27.37 20.98 24.7? 131.7? 28.3? 231
Wages 241.02 9.52 25.41 2.04 23.92 '12.1 0 16.98
Other 5.48 0.41 4.83 1.09 6.49 1 .73 3. 70
TOTAL INFLOWS 130.22 53.04 120.05. .73.12 255,29 66A3 031
Inputs .6. 14 1.7 2.16 8.56 :-. 94 4.84 4.47
Grain and Meat 14.17 6 10.46 13.01 1.? 10.11 11.38
Other Food 18.37 6.4-1 14.91 12.92 18.34 11.27 141
Livestock 3.18 0.00 3.Z3 0.00 4.73 0.18 2.06
Household Goods 21. S. 5. 87 19.73 9.44 23 34 11. 06 16.09
Other 1 .49 0.8? 11.53 0.83, 1.16 1.36' 1.28
SUB-TOTAL .64.97 30.82 52.02 44.9 6 44.60 3U.83 49:.47:
Gifts and Loans 0.92 0.46 0.95 0.42 0.93 0.6, 0.76,
Wages 3.14 0.68 1.90 3.34 3.27, 1.21 2.06
Transport 4.74 1.42 4.04 2.74 6.07. 1.81 3.57
other : 23.45 16.08 .23.26 20.22 20.90 3.5 22.16
TOTAL OUTFLOWS 9.2 39.46 82.17 70.68 95.77 65.52 78,02
MET CASHFLOV 1i 1.00 13.58 :37.88 2.44, 59.52 0.86 25.11
a Average for 3? households participating ini tha ?IVR.
fees paid for hiring tr action are* not induclded. If these were added in, it i cidz
substantially alter the net cashiflow position of hire-in households.
File: .AAVS4?l~ah.T -7.5 D fate. 13t10114

fees. For this reason, only comparisons between villages, sex of head and cattle asset categories are presented. Again, formal cashflow analysis tends to confirm informal observations. Livestock sales was the single largest source of cash income, but was obtained irregularly -- in large amounts -- at one or two ocasions per. household. during :the year. (The irregularity was confirmed via analysis of cashfiow patterns by month.) Shoshorig, 'iale, .I d rich households all received much more cash from livestock sales than their comparison groups. Crop sales provided very little cash income, as would be expected in a drought year. Beer sales provided a substantial and regular source of cash income for several households, particularly female households in Shoshong. Somewhat suprisingly, rich households had higher beer sales cash income than did poorer households.
Gifts and loans revenues, primarily remittances from non-resident household members, were an expectedly large source of cash for all categories of households. Village wage cash incomes were important, particularly in Shoshong, for male households and for richer households. Overall, aproximately P60.00 was received each month via sales and a total. of Just over P100.00 was received from all sources. One clear implication of these findings is confirmation of the need to evaluate cropping enterprises in a whole .farm context.
Turning to cash outflows, purchases of grain .and meal, other food, and household goods were regular and substantial for all types of households. The amounts spent on inputs and livestock were relatively little. Input expenditures were higher for female households, reflecting purchases of grain for making beer. Essentially all MVRU households lost little cash to gifts, loans or wages, relative to what they received Miscellaneous expenditures, primarily for various services, accounted for about F20. 00 per month.
The net cashflow per month was positive for all households, but this will likely change when traction hire fees are added in. Male and rich households have more cash transactions and higher positive net cashflows than do their compar-sion households. Shoshong is a more monatised village than is Makwate.
Much progress has been made this year with the 11iJRU. The formal, survey format. is beginning to pay dividends in terms of fewer data gaps and ease of data checking. Some outstanding concerns, however, have to be dealt with before the survey: is continued next season. Three. issues of particular concern to ATIP Mahalapye are currently being discussed.
(a). There has been little success in recording plot specific data,
particularly 'for activies such as bird scaring and weeding. It
may be useful to concentrate on whole farm resource suse through the
MVRIU and to collect necessary observations on a plot, by plot basis:
on specific plots selected for technical plot monitoring,
(b). Many findings regarding whole farm resource allocation patterns
confirm informal impressions. Also, general cashflow and labour
patterns are similar across households, even if levels of revenues and expenditures differ. Third, it is known that any substantial
increase in labour or cash inputs to arable activities will have
File: AAWS49/Mah. 7.6 Date: 13/10/84

particular opportunity cost activities which can and should be
taken into account. Although it can be expected, patterns may shift if rains are good in 1984-85, it is unlikely they will shift
so substantially as to alter the overall assessment of the
implications of a whole farm perspective for arable technology
(c). Substantial resources during the first two seasons were allocated
to farmers who were unable to plant. This was desirable during the
initial descriptive/diagnostic stages of ATIP since it is necessary to obtain an understanding of circumstances and resource flows of
poorer farmers. It remains to be seen, however, whether this
investment of resources should be continued next year.. By then
enough data should have been gathered to. address the question of
how farmers who do not plant survive. On the other hand, there is
still lack of understanding concerning what trade-offs are faced by
farmers which have both viable cropping enterprises and other
"opportunity cost" activities. It may be desirable to concentrate
on farmers who plant next season and maintain contact with
non-planting farmers through a separate survey instrument.
In summary, a much smaller sample, focussed on households likely to achieve
competitive returns in arable production would appear to be. a more
appropriate format for the MVRU, now that initial objectives have been
accomplished (or will be as soon as the complete data set for the first two
seasons have been analysed).
(2). Technical Monitoring
This activity was based on field direct measurements of production inputs
and crop responses and provided one descriptive and two diagnostic analyses. These were:
(a). A second season's description of technical: inputs .into arable
(b). An assessment att the plot level of crop responses to the
range of current farmer practices and l'cImaticavariables.
(c)' 'An assessment of field fevel 'activities as a function of farmer
resource and environmentaI parameters.
Nineteen farms in Makwate and 28 farms in Shoshong were monitored at the plot level. On many of these farms, tillageipantinog activities. were delayed until the very late planting season due to drought and poor draught animal condition Overall, production hectarage in 1983-84 was much less than desired by the farmers. Plot level monitoring was conducted on 23 1982-83 "Trials" farms in addition: to those. farms participating in the Multiple Visit Resource Use Survey The 1983-84 Annual Work Plan proposed only whole field activity monitoring for the first group of farms. This was expanded, however, to include plot values because of the need for additional plots for monitoring of less frequently grown crops (mines) and because of the limited rnumber 'of plots per farm due to the drought.
A series of summary forms for technical field data: was developed to aid the
File .AAWS49/Mah. i ?.7 Date: 15/10/84

ATIP village staff in data collection. This series included the following:
(a). Primary summary sheet: This sheet covers 58 plot designation
items, inptt items and crop establishment parameters characteristic
of all plots.
(b>. Schedule 1: Multiple traction sheet for plots in which more than a
single draught unit was used.
(c). Schedule 2: Farmer secondary tilUage input sheet for plots in
which secondary tillage was used.
(d). Schedule 3: Farmer seeding operation sheet for plots with planting
operations other than the traditional broadcast system.
(e). Schedule 4: Seed sample sheet for the analysis of all seed samples
(f). Schedule 5: Fertiliser input sheet for plots in which fertilisers
were applied.
(g). Schedule 6: Post-emergence farmer input sheet including
information on al!l weeding and thinning operations.
(W). Schedule 7:: Stand count sheet for recording quadrat sample data by
plot on crop emergence.
()'. Schedule 8: Harvest data sheet for recording quailrat sample data
by plot on harvest stands and fruit/head counts.
(j), Schedule9: Grain yield component and pest damage data sheet for
fruit/head samples collected by plot.
These forms were supplemented with field diagrams showing plot boundaries. Rainfall data were collected at 33 sites i 'the research area with data assigned to nearby plots Air temperatures were recorded in each village by, the :village-based extension: individual (T4). Project enumerators recorded plot 'input data on farms included in the MVRU survey. The T is collected this information on the remaining farms.
Information is not complete for all plots. Weeding aind other labour/time data are often missing. Air temperatures were not accurately recorded at various times in the season. In most cases, labour/time di ta is not critical to plot level diagnostic analyses. Wheni needed, typical values from the MVRU survey can be substituted for missing values. When missing, temperature data can be approximated by Mahalaoye data. It is encouraging that the village staff have recognised the need to collect temperature data correctly.
Table 7.3 gives a summary of 1983-84 whole farm cultivation activities by estimated soil moisture status for 38 ATIP farmers grouped by recommendation domain categories.
It is important to note the considerable area that is plough-planted by all traction categories when soil moisture is considered less than ideal (soil
File: .AAWS49/Mah.1 7.8 Date: 15/10/84

moisture levels 3 and 4). Soil moisture level 4 is recognised as being insufficient for planting purposes. Yet the plough-planting that does continue under this poor soil moisture demonstrates the potential that exists for tillage operations, if not planting, on a considerable number of days. This observation is the basis for proposing the sole ploughing plus planting methods trial (Section Figures blocked in Table 7.3 represent percentages of operations that were done on less than ideal soil moisture by farmers who have control of their own draught situation. Alternative tillage-planting strategies will likely be proposed for these farmers and those conditions.
VILLAGE FARMERS TOTAL TOTAL --------------------- ---------------------AREA DAYS 1 2 4 1 2 3 4
Tractor -Own (S) 3 21.4 5.0 0.0 77.1 :16.8 6.2: 0.0 81.0 :14.3 4.71
Osen- Own (S,M) 7 5.0 9.1 0.8 59.7 :39.2 4.9: 2.4 54.4 :40.8 2.4:
Donkeys.- Own (M) .7 5.2 12.7 0.0 44.5 154.9 0.81 O00 50.0 :49.1 1.1'
SUB-TOTAL 17 0.2 63.8 ;32.8 4.4: 1.0 7.3 139.5 2.3:
Animal plus Tractor 5 5.0 4.4 25.9 56.9 15.7 1.7 34.2 33.3 36.0 2.5
Tractor Hire 6 6.3 5.2 0.0 41.8 53.3 5.:4 0.0 44.4 38.9 16.7
Animal Hire (M) 5 4.0 0.0 44.1 5 .0 0.0 0.0 54.0 46.0 0.0
Coop (S,H) 5 2.6 7.2 0.0 76.2 17.6 6.3 .0.0 59.0 21.0 20.0
OVERALL 38 5.1 11.L 37.3 6.5 4.9 51.8 36.6 6.6
a ..
Planting soil mixture evaluated by farmeriresearcher: I = excessive for ope'taion, 2 = good
to excellent, 3 = sufficient but drying, 4 = no sufficient.
(3). Inventory Surveys
It has increasingly appeared that farmer assets -- human, livestock, and fixed capital -- are a ddminant,' if not the dominant, element influencing husbandry practices and management strategies. This is particularly important in the Central Region where the technical element and exogenous human element (institutions, suppott services, etc.) are similar for most farmers. Therefore, three special surveys were designed and administered to farmers participating in the MVRU to give anmore detailed picture of their assets than had been possible until that time. The three surveys were:
(a). Household Census (an update and expansion earlier census).
(b). Household Livestock Inventory-- which detailed herd contpositon for
File: .AAWS49/Mah. 7.9 Date: 15/10/84q

cattle and goats and included"a retrospective on -inventory changes.
(c). Farm Fixed Capital Inventory.
Databases have been created:for eAcsh urvey and available data have, been analysed. Partial findings were presented -it the Sebele Arable Research Review Meeting.
Summaries of information gained via these three urveys are presented in Tables 7.4 through 7.7.
Table 7.4 summarises data on household compositions and participation in arable activity. Results are presented for the same household groupings used min Table 7.1; Village, sex of head, cattle assets, and draught access. A key overall finding is that household demographics and participation in arable activities differs little across these groupings. Ihis can be
distinguished from the consistentt differences found in areas ploughed, sorghum harvested, and some types of cashflows.
VARIABLE -------- -- ------- ------------- -------------SHOSHONG HAKWATE HALE FEMALE )35 0-35 OWN DEPENDENT
Locil Residents 9.4 10.9 9.5 10.7 8.8 10.6 9.9 9.9 9.9
Percent. of Members Who Are
men _23 22 24 21 2 21 24 21 23J
Vain 22 T0 .10 23 11 17 1 2 23 21
Boys 2 31 .9 .-3L 23 29 20 26
Girls 33 26 27 35 23 35 7 36 31
Consumer Unitsa 6.6 .8.2 7.3 6.9 7. 7.2 7tS 5 7.2
Consumer UnitslResident 0.72 0.82 0.80 0.67 0.71 0.87 0.80 .0.47 0.16
S Whose Main Actiiity
IssFarning 27. 2 27 23 28 24 24 24 25
No. Active in Ploughing. 3.7 4.3 3.9 4.0 3.9 4.0 4.1 3.5 4.0
S Residents.Who Plough 27 33 31 27 34 27 33 23 0
S Residents Who Weed 21 23 24 19 25 20 24 17 2
Worker Equivalents 3.1 3.3 3.2 3.3 3.2 3.4 3.4 3.0 3.3
Definitions for these terms are given in Table 7.1.
Households average around 10 resident members. Approximately 45 percent of
File: AAWS49/Mah. 7.10 Date: 15110184

resident members are adults. Girls (under 16) are particularly prevalent in poor, female headed households. A larger proportion of residents are boys in Makwate than Shoshong, probably reflecting the fact that fewer Makwate households have distant cattle posts. Consumer units were derived using adult caloric requirements to weight individuals of different ages. Overall, households have to feed just over seven adult consumer units.
Approximately 25 percent of resident members consider crop farming to be their main activity. This is interesting when compared to the proportion of cash income received from crop sales. Three explanations are partially responsible for this:
(a). Most crop production is consumed, not sold.
(b). Perceptions of crop farming are based on long run outcomes, not
just resultsin recent years.
(c). Crop farming is a way of life, as well as an income activity. Many
individuals have no alternatives even if returns to cropping
activities are extremely low.
Approximately four people are involved iri ploiijhing -- including seeding and preparing food. Thirty percent of residents participate in plough/planting and a smaller proportion, 22 percent are active in weeding. Since not all active members are active in each major activity, a concept of worker equivalents is presented, indicating the number of people who are active in cropping during plough/planting, weeding and harvesting (113 unit given for each person active in each of theses three periods). Worker equivalents are not weighted by age. On average, and.for each group of households, just over three people per household work full time as crop farmers.
Table 7.5 presents information on fields and ploughs of. the MVRU farmers, distinguishing only between Shoshong aind Makwate households. All households have access to at least one field; 22 percent have access to a second field. The fields have been cultivated for decades. On average, the current heads of households obtained their fields twenty years ago. More than half the fields were obtained via inheritance.. Twelve percent of fields were borrowed or were obtained as gifts. Receiving. the right to use land as a gift is particularly important in Makwate, accounting for nearly 20 percent of fields -- most of those used by poor, female households. Forty-five percent of fields in Shoshong and 36 percent overall were intially opened (destumped and/or .fenced) by the current household head. A majority, of fields are completely or, partially wire fenced. The proportion" of fields with some wire fencing has inceased dramatically among Shoshong households since those households first began cooperating with ATIP. Forty percent of Shoshong MVRU fields, however, remain without any fencing:
Most households have one single-furrow mouldboard plough. Nineteen percent of households have no plough and 22 percent have two or more ploughs. Over 60 percent of ploughs owned were purchased more than 20 years ago and only half the ploughs were purchased new. Ahy depreciation value for ploughs would be extremely small for most households. Most farmers feel the shear(s) on their ploughs are in good condition.
File: AAWS49/1ah. 7.11 Date: I/I0/84

An overview of the livestock inventories of MVRU presented in Table 7.6. Cattle inventories declined slightly during the year, while goat herds increased slightly.
% Households With One 71 90 78
% Households With Two or More 29 10 22
Years Since Head Obtained 21 18 20
How Fields Were Obtained (% of Fields)
Heritance 45 64 52
Borrowed 5 0 3
Gift 5 18 9
Head Opened New Land 45 18 36
Destumping (% of Fields) a.
Completely Destumped 29 20 26
Few Stumps 29 50 .37
Part Destumped, Part Not 35 20 30
Not Destumped 6 10 8.
Fencing (% of Fields) a
Wire 24 30 26
Part Wire, Part Bush 18 40 26
Bush 18 30 22
None 40 0 26
% Households Without Plough 18 20 19
% Households With One Plough 59 60 59
% Households With > i Plough 24. 20 22
% Ploughs Purchased More Than
Twenty Years Ago 61 60 61
% Ploughs Purchased New 61 30 50
% Ploughs With Good Shear 78 90 82
a Main field only.
Death rates were quite high for cattle, around 30 percent, but this was offset -- in numbers, not value -- by births. Cattle herds comprised mainly of cows and their calves. The percent of bulls was small, but was inflated for Makwate since a couple of farmers with very small herds had bulls. Shoshong farmers lost a substantial number of cattle. Few cattle were slaughtered, none in Makwate. The overall offtake rate for cattle was a respectable 16 percent. Factors contributing to the relatively high offtake rate were the continuing drought which required cattle to be sold
File: AAWS49I/Mah.i 7.12 Date: 15/10/84

AREA, t"983-84
Numbez of Househol ds Ow~n ing !7 522
....Beginning Invenioiy 42.4 176 36.8
End in g Inventory 35.2 2 1:. 32.1
Average Inventory 8. 9534.4
Percent Calves 24.6 30125.4
Pe r jen t Cows .52. 9 48.4 52.2
Percent Oxen* 20 .0 4.6 17.7
Percent Bulls .01:7.03.
Birtrate31 3 30
Dpathrate. 30 31: 30
Percent Lost to 0
Percent Slaughtered 0
Percent Purchased This Year 2 2
Percent Sold This Yezr 23 11
Offtake Ra te *. 3
Number of Households Owning 17 9 26
Beginning inventory 16.4 8.4. 13.7
Ending inventry 25.4 7.7 19.3
Average Inventory 20.9 8116.5
Percent Kids .41.4 26.0 36.3
Percent Femtle Adults 45.1 60V6 50.2
Percent Male Adul'ts- 13.4 13.4 1.4
*Bir thrtate .. 5. 8348. 0
.De at 1.ra te 9 .7115.4
Pe i cenht L os't% 6.~ 6.0 41 .9 7. 9
P er cen t S I au g ht d -2.141 13.1
Percent. Purchased This Year I1 6 ..3.6 2.3
Percent 'tolId This "ear 8.0 2.9 4.
Offtake: xae ... 2 17.0 .19.4
NUMB ER OF CHICKEN .10 7 .. 0Z 10.5
NUMBER OF SHEEP 4.1 1 2.0 3.4
File: .AAWS49/MahT 7.13% Oate 15/10/84

to avoid stock starvation and to rcrse cash for family needs, and the fact that five farmers sold substantial numbers of cattle to invest in housing and fencing The high offtake rate was achieved in spite of the fact that the price farmers received per head during the period April/May 1983 to April/May 1984 was P180 compared with PZ50 in more normal periods
Goat herds comprised mostly adult females and their kids. Goat birthrates were higher than for cattle, and deathrates were much lower. Over 13 percent of the average inventory of goats were slaughtered during the year, indicating the greater importance of goat meat relative to beef. The overall offtake rate for goats-was nearly 20 percent, despite a small proportion of goats being sold.
Most households also had chickens. Only a few had sheep and none had either horses or pigs. One had a mule, but it died. Donkeys were, of course, more prevalent in Makwate.
Values of livestock inventories and farm f:ted capital are presented in Table 7.7. Results are again presented for the four key groupings: village, sex of head, cattle assets, and draught access. The value of livestock inventories is distinctly greater for -hoshong households, male headed households, and draught controlling households, not to mention the obviously larger value for cattle rich households. The values of goat and chicken inventories tend to be higher for housahblds with higher valued cattle inventories.
Accumulated livestock assets were not, turnedinto agricultural capital, as is suggested by the fact that households with lower livestock assets did not have less agricultural capital assets. (Other equipment can only partially be treated. as agricultural capital since it included carts and, for a few households, vehicles which had non agricultural uses.) Hand tools and receptacles represented a hi gher proportion of agricultural capital than might ordinarily be expected.
A substantial proportion of assets were tied up in buildings, particularly village buildings. This is particularly true in Shoshong. Households with more cattle have buildings which are worth more than double those of households with fewer cattle assets. Neither female households nor draught dependent households have fewer building assets. Female -households have as much housing assets as male households despite lower incomes, mostly due to inheritance, family help and remittances.
Overall, households have around F1O,000 in assets. Shoshong, male, cattle rich and draught controllers all have greater assets, largely due to livestock assets. It should be noted, these same households achieve better arable results, but do not have more agricultural capital -- except for ploughs. Livestock account for just over 6-0 percent of assets overall and buildings for another 32 percent. Agricultural capital, particularly if one were to eliminate vehicles and carts,-, represents an insignificant proportion of household assets. . "NEW VILLAGE" DESCRIFTION AND DIAGI1ZSIS
ATIP Mahalapye had hoped to begin research in a third. village, Makoro in the Palapye District, during the 1983-84 season.. it was not possible to
File: .AAWS49/1Mah. 7.14 Date. 15/10/84

Cattle 6225 344? 7101 21324 0040 132 .1V 087 !264
Coats 488 154 398 332 516 267 384 356 373
poultry 59 51 41 50 71 41 5? 45 57
Donkeys 9t 260 2W 65 124 173 243 2 53
Other 172 66, 76 C. 8 43 7? 395
TOTAL 7041 S180 7063 2771 &0872 2256 7644- 352 5906
Ploughs to 14 131 33 144 614 135 .35 94
Other Equipment. 053 284 204 536 15 5 446 Zia 464 328
Hand Tools 53 .50 1 46 75.:. 41 55 56 55
Recepta cles 12 96- 115 to& 131: ?? 12 97. 112
TJOTA L 301 70 si1 7231 5.0S 6461 546 652 58
L4adseArea !06 153 484 202. .64 195 484.. 226 379
Village 3594 .1020 2n127 170 460.o, 17 06 2524.8 2791 2640
TOTAL ASSETS 11742 5733 iU8 11a '8Y-4S6 140,24 4303 122 703 1514
Livestock .60 67 71A 2 LS 47 .68, 4? 62.
Buildings 34 21. 2 60 29 39 27. 42 32
Agricultural Capital 5 10 5 11 ~ 14 5 9 6
Valuation of ploughs: other equipment, .ind buildiu~s based on farmet s''assessnriets tO
current cost to obtain~a repiscement in "he sam~e condition. Livestock valued as follows,
bull =P3lo, cow =P20S, Olen =P220, calf P40, adult goat = P2, i 1, sheep P28,
thickens =P5, donkeys = P65, and mules = 170. ThEse are conservative estimates and
-represent the minimum value of livestock inrenteries. Hand tools were valued as follows: hae
P4; ate, pick, spade, iake =PIO; wheelbarrow =P25, auger =Ft. Receptacles were valued as
follows: small basket =P2.50, large basket =P6.00, small bucket =FSA.10, large bucket
P9.00; drum P16.00; plastic container P15.00, sack P1.75, large bowl =P3.50, small bowl
File: .AAW549IIHah.T .5-Date: 114184

intiate field trials due to lack of staff from DAR and DAFS. In anticipation of beginning trials research next season, an exploratory survey was designed and carried out in May and June. Additional exploratory visits to Makoro likely will continue.
The exploratory survey format used in Makoro differed from that used earlier-in Shoqshong East and Makwate, reflecting greater knowledge of farming practices in the survey area: The survey design required field teams to fill-in a "Single-Visit Farm Characteristics and Cropping Outcomes" form. On this single sheet, objective information on household composition, agricultural resource position, livestock management, and cultivation activities during the 1983-84 season were recorded. Following recording of information on a farmer's circumstances, a series of subjective questions were asked. During the first two days of interviews, subjective questions fell into the following four categories:
(a). Farmer priorities.
(b). farmer assessment of arable production conditions.
(c). Knowledge of and interest in DAFS recommendations.
(d). Responses to multiple tillage.
Information in the first two categories was difficult to pursue, in part because it was the end of the cropping season and questions seemed too abstract to farmers. Questions about the last two categories met with rapidily diminishing returns since essentially none of the farmers were familiar with DAFS recommendations nor had observed multiple tillage.
In a debriefing session held after the second exploratory visit, the team decided to shift the focus of informal questioning to the following issues: problems and priorities of permanent lands dwellers versus seasonal migrants; settlement patterns as related to nearby villages and definitions of lands areas; and draught arrangements.
A distinguishing feature of Makoro is it is not a village, at Teast in the normal settlement pattern. Households in Makoro might have village residences in Serowe, Palapye or Radisele. 4:any have -no -village residences and are permanent lands, dwellers. Permanent lands residence is possible due to three water pans which generally do not dry during:-winter, even though there is only one borehole serving the whole area. Makoro also is of interest because there ii only one small-trading store in theA area even though it is adjacent to the Gaborone to Francistown road. Tractor hire is the main source of draught, closely followed by oxen.
In summary, the settlement pattern of lMakoro has presented ATIP Mahalapye with a distinctly different research environment from that found in Makwate and Shoshong East, even though cultivation practices are quite similar -particularly -in relation to Shoshong East. Furthe r exploratory visits (to Makoro and regionally) will allow evaluation of the significance of the "Makoro pattern" in other parts of the region.
The planned sample frame census for Makoro will not be administered. This decision is based on the following.
(a). Farmer selection for either trial research or the MVRU survey will
be according to recommendation domains. It is clear on the.basis
File: .AAWS49/Mah.1 7.16 Dae: 15/10/84

of the exploratory survey that the dominant RD in .Makoro is tractor
hire, followed by use of owned cattle. We feel information from
the exploratory survey can be used to select ATIP farmers in
(b). The absence of a village center in Makoro means that a proportion
of farming families are actually residents of other villages.
Also, within any given household, some members may be resident and
others seasonally migrate to villages. Third, the Makoro lands area merges indistinguishingly with other lands areas. Taking
these factors into account, it is not clear there is a distinct
"Makoro population" and therefore reason for a sample frame of that
(c). The extent to which ATIP Mahalapye will be active in Makoro still
is under discussion. Given planned RM,RI and RM,FI research in
the Mahalapye lands areas and cooperative demonstrations with AD's
representing several extension areas (see Section, Makoro
may serve only as a FM,FI testing village rather than a primary
research extension area (as Shoshong East and Makwate). Also, ATIP
Mahalapye has not yet been assigned a third T4. COWPEA BASELINE STUDY
Prior to the 1983-84 season, DeMooy, the Cowpea CRSP Agronomist located at Sebele, asked ATIP to design and administer a study on the role of cowpeas in Batswana farming systems and cowpea husbandry practices. In February,- a cowpeas baseline survey was cooperatively designed by Miller eBaker and Norman. The survey intially was pre-tested and adminstered to ATIP farmers in the Francistown area. Additional pre-testing and modification was carried-out in Mahalapye. The survey was administered to.49 ATIP farmers. in Shoshong East and Makwate during June. The survey was administered ,by the three field enumerators, the two T4s, Tibone and. Luzani i
Data from the survey has entered on the Sebebe computer. Data analysis has been completed. Summary results have been compiled and are available for distribution in ATIP Progress Report M84-5. Some of the key. findings from. the survey are presented in Tables 7.8, 7.9 and ?.10.
Table 7.7 presents an overview of cowpea production and production practices in Shoshong and Makwate. Nearly 70 percent of households have grown cowpeas in each of the past 5 years; a smaller proportion in Makwate due to drought and subsequent lack of seed. Only eight percent of households have grown cowpeas less than three Vears in the past five. The primary reason for not growing cowpeas was a lack of seed. Twenty-four percent households did not plant cowpeas: or anything else during the past two seasons due to drought.
Draught used for cowpeas was the same as for other crops: primarily cattle and tractors in Shoshong and donkeys in Makwate. Nearly all housholds broadcast seed. Floughing was the only soil tillage, except .for some:, households harrowing in Shoshong. Eighty-eight percent of households-grow cowpeas in mixtures but. 41 percent planted at. least some cowpeas in sole plots. Cowpea mixtures nearly always included sorghum and melons. Cowpeas were mixed with maize by 31 percent. of households and were mixed with other
File: .AAVWS49/Mah. I 7.17 Date: 15/10/84

VARIABLE - -- - - - - -
All 5 years 77 67
3 or 5 ears 13 4Z 25
2 or less years 108
Lacked Seed 50 86 64
Lacked Labour 20 0 12
Did No Planting 320 14 24
Donkeys 3 .. 79 3 3
Cattle 37 21 30
Tractors so 0 31
Tractors and Animals 10 0 6
Cowpeas Sole 6.*12
Sole and Mixed 23:729
Mixed Only, 737 59
Mized-with S orgqhum'6 53 57
Mixed with Mia ie 3721 ..3!
Mixed with Melons 97 74 88
Mixed'with O thfer Beans.. 27 21 25
Blackeye 4/ 79 -59
Tswana 87 8.9
5 Kgs. 57 79 68
5 5Kgs. 53< 100 70
.5-10 llgs. 48 0 30
Did Not Weed .I
Weeded Once Y7 79 90
Weeda-d. Once 010 4
WHEN. NEEDED. 70 84 .7.
Picked All-Leaves 17 32 2
Picked Trailing Vines 17 0 10;
Picked Some.Leaves -.6 861
No Strategy 3 06
aOf households which did not grow,
File: .AAWS49/Mah.T -7.18 -Date: 15/1184

VARIABLE ------------------------------- -- --SHOSHONG IMAKWATE ALL
Had Difficulty Getting
Good Stand Establishment 77 74 76
Problems with Stand
Too Few Plants 30 16i 25
Stand Uneven 33 32 33
Stand Too Dense 10 21 14
Observed Serious Feeding 97 90 94
What Damage Observed :
Leaf and Stem Feeding 60 68 63
Flower Feeding 50 58 53
Pod Feeding 57 79 -. 65
What Did If Insect Problem
Generally Nothing 97 84 92
Ever Spray for Insects 0 0 0
Bird Damage 30 32 31
Other Wildlife Damage 90 90 90
Serious Disease Damage 60 90 71
Which Type Observed
On Pods 17 68 37
On Leaves and/or Stems 47 74 57
What Did If Disease Problem
Generally Nothing 63 37 53
Took Seed Only from
Healthy Plants 3 53 21
File: AAWS49/Mah.T 7.19- Date. 15i1/84

VARIABLE ------------------- ------------------SHOSHONG MIAKWATE ALL
Eaton By Household 20 12 i7
Eaten and Save for Seed 53 53 -50
Eaten, Saved, Sold 27 35 30
Eaten by Household 70 89 78
Eaten and Sold 30 11 22
Ever Sell Cowpea Grain 60 63 61
To Whom Sell
Another Farmer 67 43 ,,8
Local Trader 25 4 32
BAMBICoop 8 14 10
Ever Sell Leaves 37 2z1 31
How Often Post-Harvest
< Once a Week 10 74 35
Once a Week 27 5 1.8
> Once a Week 63 21 47
How Often Winter
< Once a Week 23 79 45,
Once a Week 33 0 20
> Once a Week 43 21 35
How Often Summer
( Once a Week 40 84 7
Once a Week 30 0 18
) Once a Week 30 16 25
Household Production Z? ".. .. 31
Other Farmers 1 7 5 12
Local Trader 3. 11 6
Household and Traders 23 16 20
Household, Other. -- mers.........
and Traders 30 31 31
< Once a Week 7 47 22
) Once a Week 93 53 78
Household Lands 33 53 39
Other Households 17 16 17
Lands and Other Households 50 32 43
File: .AAWS49/Iah.T 7.20 Date: 15/110/84

beans by -only 25 percent of households.
Blackeye and Tswana were essentially the only cowpea varieties grown. Blackeye was grown by 5? percent of households and Tswana beans by 88 percent of households. The amount of seed planted was very small, under five kgs. for 68 percent of households. No households planted more than 10 kgs. of either variety.
Most cowpea plots were weeded once. Nearly 80 percent of households did not believe labour was lacking to grow cowpeas. Several different leaf gathering strategies were used.
Farmers believed they faced severe husbandry problems in growing cowpeas, as is shown in Table 7.9. Three-quarters said they had difficulty in establishing good stands, 8Z percent had observed serious flower drop, 94 percent had observed serious insect feeding, 69 percent believed there were special weed problems, 90 percent felt wildlife damage was a problem, and 71 percent thought diseases were a problem. Despite farmers' concerns, farmers generally did nothing with regard to eiher pests or diseases.
Table 7.10 summarises information on cowpea utilisation. Most cowpea grain harvested is eaten or saved for seed. Twenty-two percent of households sold some grain the last yeac they harvested grain. Sixty-one percent of households reported having sold grain at some time in the past. Most grain was sold to other farmers, followed by sales to local traders. Thirty-one percent say they have sold cowpea leaves.
The frequency with which cowpeas were consumed varied with the season. Immediately after harvest, 35 percent consumed grain less than once a week but this increased to 57 percent during summer, preceeding harvest. Cowpea grain consumption was less frequent in Makwate than in Shoshong. Households obtained cowpeas from a variety of sources, including household production, other farmers, a"id local traders. Seventy-eight percent of households ate cowpea leaves more than once a week during the post-harvest period. Cowpeas leaves were obtained from household lands or from other farmers. REGIONAL PERSPECTIVE STUDY
The regional perspective study has been delayed until the 1984-85 season. The objectives, justification and approach to regional perspective studies will be different in the coming season than had been planned for this season (see Section The regional perspective study intially had several descriptive and diagnostics objectives, some to be accomplished through a regional exploratory and some via a regional farm management survey.
Most of the objectives envisioned for formal survey, those relating, to a description of. farming practices throughout -the Central Region, were accomplished through analysis of the 1983 Agricultural Demonstrator survey. In fact, results of the AD survey suggested a multi-stage sampling procedure, logistically necessary, could lead to unrepresentative results. This is because practices tend to be relatedd to whether a village is a donkey, cattle or tractor traction dominated village. A stratified mul ti-stage sampling approach toud be used, but then ATIP Mahalapye
File: .AAWS49/Mah.1 7.21 Date: 15/10/84

already have descriptive information relative to a representative donkey and a representative cattle-tractor village (Makwate and Shoshong). The representativenss of the ATIP research villages was confirmed n the AD survey, eliminating another purpose of the formal regional survey.
The planned regional exploratory survey will be conducted in the coming season. As is explained in Section 7.2 1.3, the objective has shifted from system description to a focus on why certain patterns of draught power and progressive practices are observed in particular areas. In general, ATIP Mahalapye feel emphasis should shift from description to monitoring patterns of technical change. DESCRIPTIVE/DiAGNOSTIC WORK NOT MENTIONED IN THE
1983-84 WORK PLAN
(1). CropPing Plans Survey
This survey was designed to provide three categories of information:
(a) Planned traction use, crops to be grown and seed availability as
needed to decide which farmers might participate in the cowpeas
seed comparison and seed treatment trials.
(b). Expected use of houshold and non-household resources to help check
for gaps in technical and resource-use monitoring data.
(c). Farmers' early season plans to compare with actual activities as
the season unfolded.
The survey was administered during 11th-17th November to 45, of the 52, ATIP farming households in Shoshong East and Makwake The data were immediately entered in the Sebele computer and analysed. Results were used, as planned, in selecting cooperators for the above trials. A brief report on survey results was prepared and circulated in December. The report, ATIP Progress Report M83-2, is available through ATIP Mahalapye.
A major finding of the survey was that ATIP farmers started the 1983-84 season in a difficult position: most were short of seed and had animals unfit for traction. The government seed distribution programme, provided a large proportion of the sorghum seed farmers had on hand (78% of households received seed). The programme also led to .more farmers planning to plant more maize than normally is the case.. Maize is not a successful crop in the Mahalapye area. In general, donkey farmers were able to start earlier, despite worse grazing in Makwate. Many farmers in Shoshong planned to shift to tractor hire since their cattle were not fit, an option that is not available in Makwate. The shift to tractor hire partially appears, to reflect a long run trend toward a multiple traction -- tractor and cattle
-- system.
(2). institutions, Services and Infrastructure Survey
This survey was designed to better describe the exogenous human element influencing farming practices. The questionnaire for the survey was open ended. It was administered in Shoshong Last and Makwate. Questions on
File: AAWS49/Mah.2 7.22 Date: 17/10/1984

government programmes, health and education facilities, government officers, group activities, and availability and prices of goods in stores were asked by two BAC students assigned to ATIF Mahalapye during their January and February break. Questions on cattle posts, beer brewing and providers of village services were asked by temporarily hired village residents. USAID technicians and their counterparts asked questions about village leaders and key institutions such as the Land Board, Farmers Committees and the Village Developmnt Committee.
There are no current plans to present the results of this survey in a seperate report. The rural sociologist joining ATIF for the 1?84-85 season, Lesotlho, is currently reviewing the survey results and will carry out follow-up visits where he feels it is necessary. Information on prices will be used in analysis of the MVRU data.
(3). Weed Growth Survey
The results of this survey provide a quantified description of the range of weed problems encountered in the 1983-84 season. This quantification should also be useful as a response variable in analyses of farmer and environmental (soil and climate) inputs that contribute to weed control or weed problems within the currently practiced low input system. Pertinent farmer activities on environmental inputs included depth of ploughing, moisture at ploughing and the raifrifall profile.
Cultivated 0.75 3.14 0.71 4.04
Uncultivated 1.00 7.89 1.00 8.00
Cultivated 0.92 7.54 0.45 4.Zo
Uncultivated 1.00 15.16 1.00 5 80
Cultivated 0.75 1 17 1 00 5.23
Uncultivated 1.00 4 75 0 88 2.13.
Cultivated 0.67 0.95 0 92 3.80
Uncultivated 0.38 1.18 0.88 3.88
a Proportion of samples infested.
The lack of farmer response to extension weeding recommendations suggests that farmers have concluded that investments in clean and early weeding would not generally payoff. 'At present,. most. farmers::weed only once and. maybe not at all on most of the area planted. Weeding generally begins
File: .AAWS49/Mah.2 7.23 Date: 17/10/1984

well past the stage of crop growth at which OAFS currently suggests it is most beneficial to crop yield.
The survey was conducted by project techni.Cal staff. Table: 7.11! shows the .extent to which weeds were a problem in. mid iind late season on farmer's fields in the ATIP research area in 1983-84- Assessments were made prior to weeding. Rainfall limited weed growth throughout the season as indicated by the small percent weed coverage on even uncultivated plots.
WEED SPECIES-- - - - - - - -
-------------------------- ----------------SHOSHONG MAKWATE SHOSHCNC NAKIVATE
Ipomea spp. 0.25 0.43.33 0.33
Convolovulus arvensis 0.57. 0.45 A.00 3.33
Curcubitaceae 10,53 0.25 0'.6 1.33
Xanthium spp. 0.25 .1.33 4.66
Cassiz italica .0-28. 0. 50 i .00
Datura spp. 0.33
Amaranthus spp. -- 1.00
Solanum spp. 0.28 -2.00
Other Annual Broadlea&f 0.3'5 -2'.66 0.66
Striga asiatica .3 0.66
Sida spp. .0. 75 1 .66 z 00 0.33
Herbaceous Shrubs --1.66 0.33
Urochloa spp .11.62-Eleusine indica 0 .07 0.45-Cynodon dactylon --06
Rhynelytriuu& repens 0.10 1 .16
Chloris-virgata 0.28. 0 .75-Eragrostis spp. 0. 141 0.75 1-.66 1 .66
Panicum maximum -' 033
Tragus berteronianus 0.87 0.90 0.66 0.33
Aristida spp. 0 39' 1 .0 1.33 1 .00
Digitaria spp. 0.28 0.75 1 1.00
Enneapogon cenchroides -02
Table 7. 12 gives a listing of' .the major, weed species, found: in fields of the research area.
Weed control levels were asses-sed, on project:- tillage. trial plots '(section
.1.1), in 1983-84. Results, presented I ale?1j so ha al
.Ploughing plus planting operatio ns gave better weed. co:ntro-l than did the
File. .AAWS49IMah.2 -7.24 -Date: 17/1011984

traditional broadcastlsingle ploughing check. Broadcast-second ploughing was the best planting operation in terms of weed control. ATIP Mahalapye believe that the row planted plots suffered more weed growth than is indicated by these data. This is probably because the late season rains produced weed growth that was particularly noticeable in the row plantings after the weed assessment was made.
(4). Soil/Root Profile Survey
Soil/root profiles to 80cm. were analysed on specific micro-topographical sites on seven *sbrghum fields in Makwate and on nine sorghum fields in Shoshong. This undertaking was supported by i1982_83 .ATiP soil. test data that 'show no correlation of furrow slice soil analyses with plant.growth and by indications from Sebele and from South African extension .that subsoil charactersities can be limiting to root development in soils of this region. This survey was further motivated by observations that sorghum root development appeared to, be far less than would be indicated by even the poor rainfall pattern experienced these past two seasons. Subsoil characteristics considered for their effect on measured root development included soil pH, exchangeable aluminium, available phosphorous, and compaction layers.
TRADITIONAL CHECK 1.00 0.88 11.33 6.50 27..83
------------------------------- ----------------------------EARLY PLOUGHING FLUS:
Row Planting 0.92 0. 7:9 5.79 5.08 10..87
SHa:rrow Planting 0.88 0.79. 6.16 4.66 10.82
CuIlt. Planting 0.83 0.167 6.37 4--8 10.95
Double Ploughing 0.79 0.63 5.20 3.16 8.36
a Proportion of samples infested.
Due to limited resources, many important local sorghum iandraces are 'not being improved by the breeding programme at Sebele nor increased by the Seed Multiplication Unit. Thistrial evaluate the potential of developing "best seed" types:from within several of these local.' populations. Seed samples from "69 different farmer seed lots were grown out in a replicated nursery at the M.ahalapye Research Sub-S'tation. These samples represent eight identified sorghum varieties plus several samples of unidentified local red sorghums.
File: .AAWS49/Mah. 2 7.25 Date: 17/10/1984

Drought conditions following planting made any chance for good information seem unlikely. Exceiient late rains in March, though, produced good results in many of the lines through late maturing head development or through tillering. Al lines were assessed for maturity, plant height, total grain yield and yield components, and stover produced.
Following the stover harvest, the stalk base was left undisturbed so as to be able to assess spring regrowth in these lines. 05D sorghum has demonstrated some capability for fallispring ratoon regrowth and grain production in the Mahalapye area. Other lines show considerably less ratooning capability. Such a ratooned crop of a well established autumn stand, would hold considerable advantage as part of a farm enterprise that experiences endemic difficulty with stand establishment and planting sufficient land- area.
The proposed field operation was not implemented in 1983-84. Post-emergence harrowing was proposed as an efficient scheme for thinning large areas with pockets of excessively high plant stands of .a cereal crop. Because plant spacing is often highly erratic w:tli"traditional plantings, measures that might be used to correct this need to be considered. There might also be some measure of early weed control with the harrowing operation.
One farmer did agree to use his own donkeys to pull a harrow over portions of a dense sorghum stand.. The farmer, though, never did manage to have his animals collected at designated times. The same farmer was quite cooperative in the implementation of two other trials on his field.
This experience points out a need for ATIP to have resources, such as draught animals, under project control for the implementation of trials and demonstrations that farmers may be reluctant to try. Farmers are struggling to produce a crop and, in cases of design activities, the outcomes may not be well understood by the researcher. Consequently six donkeys have been acquired by ATIF and are being looked after by a farmer in Makwate. RIDGE PLOUGHING
Because the ridging ploughs; arrived late in the season., early season ridge planting of maize and jugobeans was not done.. There was considerable interest shown in the equipment and if rains .Ahad fallen. during the January-February period, they could have been used for broadcast/replanting failed plots. EFFECTS OF EARLY TILLAGE ON' TRADITIONAL BROADCAST/ FLOUGHINIG
Early tillage operations were carried out by tractors hired by the project. Treatments included an early 10cm deep ploughing, an early 20cm deep ploughing, a spike-tooth harrowing and no early tillage. Strips of these treatments were cross broadcast/traditional ploughed by the farmer on his normal series of tillagelplanting days. There were two replications on two fields in Shoshong.
File: AAWS49/iMah.2 7.2i Date: 17/10/1984

Even though data wiIl show there was a significant benefit from early ploughing for stand establishment and, in the case of one field, for grain yield, benefits were not economic in this dry season. Harrowing had a detrimental effect on surface soil structure in both fields -- one sandy loam and the other a sandy clay 1oam with pockets of sand. There was no significant difference between shallow (10cm) and deep (20,cm) early ploughing.
Difficulties in implementing this trial were associated with a lack of control over the Fequired draught power. These difficulties include the timing of the early tillage operations and the depth of the early ploughing treatments. Proposed access to Mahalapye Research Sub-Station tractors will greatly aid these types of RM, RI design trials.
(1). Dual Purpose (Grain/Hay) Evaluations In Tswana Cowpea
Several methods have been proposed to improve the quality of crop residues for feeding animals. These include cutting at the earliest possible stage of growth (stooking), undersowing cereals with fodder crops, or late season plantings of sole fodder crops. ATIP has also been interested in developing the potential of currently grown crops for duaJ purpose (grain/hay, production. Tswana cowpea had already demonstrated its vegetative growth potential as a, late undersown crop in 1982-83. It was assumed that considerable v.ariation would e Kist among indigenous lines in their ability to produce both grain and hay during a full season's growth Such lines would be an important contribution to the scheme discussed in Section 7.1.S.2.
Thirty-two line selections were made at Sebele in t982-83 from a nursery collection of more than 250 Tswana cowpea populations collected by the Cowpea CASP Team at Sebele. Line selections were based on, -in order of importance:(a). Grain yield (per plant)
.(b). Vigour inden (plant canopy height and width)
(c). Erect or partially erect plant type
(d). Determinant/partial indeterminance Geaf re'.ention)
(e) .. Large 'leaf size proxyy for high leaf-stem ratio)
The "ideotype"' was thought to be a line with partial indeterminancy to improve leaf retention late in the season, good grain production and a semi-erect growth habit to facilitate cutting for hay. Lines were grown in replicated two cow plots at the Mahalapye Research Sub-Station. An adjacent hectare plot of sole Blackeye cowpea was sampled for varietal comparison with the test lines.
Data on grain yield, post grain harvest hay yield and stand establishment (plant counts) are presented in Table 7.14.
This trial was planted :n early January on adequate soil moisture. Drying conditions in late January and early February plus variable seed quality resulted in variable establishment of the plots. Establishment was
File: AAWS49/Mah2Z 7.27 Date: 17/l0,/184

8047 1129 2.9 7.
B163 87Z 2.7 13.3
BOO7 F 779 1.3 25.6
B012 D :7 33 Z.1 27.8
B162 549 .715.6
B123 535 3.9 33.9
B020 E 522 2.7 31.7
BLACKEYE 455 0.8
8008 D 444 6927.8
8105 437 .63.3
B102 388 2.8 23.4
8127 350 .126.7
B161 34 6 082
B003 C. 339 1.8 .18.9
B013-D 321 4 "5 42.3
8 00:4 A 313 S. 7 31i.1I
8122: 309 7. 0 .29.5
8117 302 2.9 32.2
B077 280 2.7 248.9
11160 5 1_4 15.6
:8013 C 243 8 .2- 3.
8 17?7 240 63.
8020 F 229 3. 1 27-.2
B029 221 4.6 2.
8027 179 5.9 25.0
8064 178 2.2 15.0
B022 158 4.7 40.0
8164 133 0.7 1.
B054 120, 0.9 10.6
8135 76 5.,3 3.
8042 59 4.7 40 0,
8171 so 3.2 .9
----- -------------------------------------------------------------Standard Error 307 1.15.
FILE: .AAWS4?/Mah.T -7.28 -Date: 151101"1984

followed by excellent, rainfall through the month of March, Compensatory growth for gaps in plots was excellent.
Even with the good growing conditions described -above, grain yields and hay yields of certain tswana Iines 'were impressiv,.e. As anticipated, the vegetative growth of most indeterminant tswana lines was far superior to Blackeye. However, for each line, the hay harvest was delayed until after three pickings of grain. This harvest pattern was followed intentionally because it simulated a probable farmer preoccupation with maximising off-take of food for humans. Hay quality was, therefore, greatly affected by lignification of stems and leaf dropping.
Selections that best demonstrated the dual purpose potential will be tested again in 1984-85 at Mahalapye and in the ALDEF fodder programme. Hay harvesting Will be earlier in order to improve hay quality. Effect of earlier harvest. on. grain' yield will be determined as well as its compatibility with traditional farmer interests. It is not yet at all certain that any fodder recommeAndition requiring special harvesting (standard residues, dual purpose residues or specta: fodder crops) will find favour with farmers in: the Cet. I Agricultural Regioin.
(2). Bird Scaring Scheme
One thousand metres of a polyethylene ",Hum Line", advertised to be effective in protecting vulnerable crops from bird feeding, was bought by the project in mid season. Bird feeding was devastating -on some sorghum plots in 1983 and is generally indicated as a major pest proberA in the project area. Opportunity costs for bird scarers '"s probably quite low but their effectivenss can also be'. limited in large fields. An inexpensive gimmick might pay for itself in bags of grain not lost to birds.
The line must be stretched taut near the top of the crop canopy, between poles spaced about 5m apart. Vibrations in the wind are to discourage birds from remaining in the area. Smalt blocks of this "protection", were installed by the project on two fields in Makwate on early maturing sorghum at mid-season. Farmers indicated that the two mraost serious grain feeders, quelea and :doves, did seem to be driven away from these protected blocks. Other birds continued to feed. The tape cracked and tore in the hot winds of this period.
Later season blocks were installed on a lar ger scale in one field in Makwate and one in Shoshong. These blocks were.approximately 0.25 and 0.50 hectares respectively. Again the birds appeared to prefer non-protected crops but nowhere was the pressure of bird feeding as intense as observed: the previous season.
It is not likely that this scheme would be practical or. economic for.. traditional farmers in the project area. Following the recommended' installation requires about P1.-20 per hecta,-re. The tape is not reusable' as was indicated because it tears easily when not taut and is very susceptible to animals passing through the crop. The project attempted to use locally obtained poles for the installation but the quality wasn't adequate for driving them firmly into the soil The purchased poles used would add considerably to the cost of the installaton.
File: .AA'S49/Maht.2 7.27 Date: 1I10/I1984

(3). On-Farm Expoerimentation Methodolocties Study
Following the suggestion of ATIP Fralcistown, five traditional plots in. the Mahalapye area were "over" samopled for harvest stan-d counts and grain yield to estimate the accuracy of the currently used data ool'iection procedures. ATIF Mahalapye agrees that because oili-fa .rm experimentation is relatively undeveloped iri terms of techniques anid statistica: guidelines, projects such as ATIF should devote s;ome Iof their resources ta provide solid experimentation ruethodology in the intere-st .of future on-farm work.
T wo. sampling Con gqurafions were tested including o-na plot consistent with the design being in'.estigated in the. ATI? Franctistown area.
7. 1. 31 .IEFFET TVFAE&S .0F S CLE F L 0U iH.I",
If recoinuiended pilantIng 71 ithoUdS are 7no1 to COLp'ete- with the ploughing, activity for the i1imited, ldeal planrting moisture periods, than a stzateqy must be investigated that would P.locate "moisture'condiltion" periods for planting and fcn iiie.partici pa ling farviers were asked to
plough "acres" without plantiria either early in "he sez'son or during drying periods in mid-planting sas*.on. lojn nilrt of -at lstona
planting -rain, four .4liiting methods. were .compared oni thims pLoughed acre. The four methods werme row planting, b!ha rrowing,
*broadcast/cultiv trig and btroa-dca:5t/se,7orld pil.:ughing. Two check comparisons wvere made. '"he .flrst &vas with a trAditional r.datsnl
*plough acre put in on th iiita a fpog~g h second Was, a
traditiO11,1l PLAnI1t On: he trttat~vnt p lanting dV
The objectives of. this test were to0
(a). AsseFss th1,e e ffec .i v ar. es 6f sole pl-oughing to imp r yve, thle erl
timingJ Irf ro e iat ionfs within ljinited draught sItuations
(W,. A s swss the ot ?-,it i-a f..;r _n.mamdi the area .illed Axnd paW
throxigi ploughing unde; 'marginal soi' conditiotiS.
CO) EvalttAte the benefits of soil tillaqe done prior to the planting
rain for crop emergence, counts, untifo:rmity and vigor of stand.,
Nine replications of this trial viere impl--mented in 11-3wate Yils were achieved on only sixr of these replications. ,Additional replications were pegged in SftOshong but never implemented because of the dr-,ought. The project was not able to smrctly enforce imp~leme.-iation of treatment plantings; only a~n good racisture. '"his was unfortunate in the sense that a strong t est of the intandpd design was not' achiev ed. It did, however, lead to some interesting oTbervations on the adatgsof the different planting methods relative to, different soil moisture at planting. Test results for grain yield hind stand est-ablishment coamparisons are given in Table 7. 5.
Results indIcate An important advantage in early ploughing. and certain planting methods for grain yield and stand estabIA-shrnent over both traditional checks. This supports the 1982-8:3 results reported in the
'ile: AAWS49/ ah2 7.0Date. 17/10111984

first ATIF Annual Report. The "economic" advantage of early ploughing is doubtful with such low overall yield levels. Three planting methods were giving good results. Only the broadcastfharrow planting failed to produce plot yields that compared well Row planting and broadcast/double ploughing gave the highest grain yields and are seem most likely to fit the resource situations of area communities.
(KGlHA) (X1,000 PLANTS/HA)
1: Traditional At
Early Ploughing 53 22 be 7.33 C
2: Traditional At
Planting Comparison 24.08 c 9.18 c
-------------------------------------------------------------EARLY PLOUGH PLUS;
Row Planter 103.96 a 11.78 bc
Broadcast/Harrow 45.64 C 8.80 c
Eroadcast/Cultivate 8154 .ab 18.53 ab
Broadcast/Double Plough 117 40.a 20.53 a
-------------------------------------------------------------"MODIFIED" EARLY PLOUGHING AND
Means in the same column followed by at least one common letter
are not different at a 0 05 level of significance.
Comparisons with traditional check are important. Check I. days were supposed to be dry. Comparisons should, therefore, show an advantage to sole ploughing in preparation tor quick.planting: following a rain rather over traditional plough-plahting on hat day.
Observations on comparisons of planting-methods suggest that tow planting, probably because of its relatively shallow seed placement, is demanding of good planting moisture in order to achieve good stand establishment;. Broadcast/second ploughing, and to a lesser extent, broadcasticultivating can produce good stands in good soil moisture and may do better than row planting under drying conditions. They also provided a much better tillage weed control than cold the row planting. DRAUGHT TEAM MANAGEMENT DESIGN SCHEME
This work might be more appropriately considered as a mixture of design/testing type work. It waS proposed that a minimIal labour investment by farmers in harvesting and storing the previous seasons stover might. yield substantial benefits if a modest quantity of feed was fed during the
File: .AAWS49/Mah. 2 7.3. Date .17/10/1984

ploughing period. Providing energy supplements on ploughing days as a
complement to existing grazing might prove attractive to draught team
owners even in years with relatively good grazing.
Available crop residue (wheat straw; and dicalcium phosphate were offered
to three donkey teams in Makwate and to one oxen team in PMakwate and one in
Shoshong just prior to the beginning of the 1985 ploughing period.
Available grazing was extremely limited and animal condition was generally poor. Nonetheless, the wheat residue not of sufficient palatablility to test the above hypothesis. Both owen teams, and eventually their owners, refused the residues. Donkeys did feed.but not in Jarge quantity. The project continued to provide mineral supplement to the donkey teams for several months. This free supplement was discontinued When the owners indicated they, were no longer interested in the programme. In at least one case, it was discovered the farmer did not tend his donkeys regularly away from the ploughing period.
ATIP Mahalapye isnot of the opinion that area farmers view supplemental feeding of draught animals as a needed activity in most years, It Is also possible that draught animals would not be the farmers,-first choice of animals to receive supplement when it is available. lIt should still be valuable to design a feeding scheme that is manageable and provides adequate feed stuffs to determine the response of farmers who are most diligent in striving for earliness in their field operations and in increasing areas tilled per day and per season EVALUATIOi OF NITrROGE (I ) AND FHOSPHO-ROUS (PF) EENEFITS
In general, ATIP MIahalapye does not feel that research data in Botswana support the testing of chemical fertilisers within the context of the traditional broadcast/ploughing system for sorghum. It is noted, however, that 20kg P/ha are recommended by DAFS when management is at a reasonable level. It was hypothesised that early ploughing -would improve soil moisture management, thereby increasing the likelihood of i crop response to fertiliser
Two replicationo5 of a strip block of single superphosphiteh were applied by the project at the recommended rate, before the initial ploughing, on 13 sole ploughing plots in Makwate and Shoshong village areas A tillage-planting sorghum trial (Section 7.2.3..) was planted on.eight of these plots in Makwate village. The Shoshong plots were early ploughed but lack of rain prevented the farmers fr6ma planting Of the eight planted plots, some g.rain yield was recorded on five On these five plots, effects of tillage-planting systems tested were signtflcant but phosphate fertiliser effects were only slightly noticeable at the highest y eld levels. Fertiliser plot sub-sampling to determine the phosphate benefit proved inconsistent. and similar estimates for control plots of the same tillage blodk were only loosely related to a more precise whole tillage-plahtiiq plot measurement. Nevertheless, farmers pi~ticipating in a project 'workshop were .mpressed with what they perceived as asignificant fertilise.effect on two fields. T'abile 7.16 gives yield estimates :or phosphate treatments in two tiliage-plianting blocks in the single best yielding replicate observed by the farmer workshop.
Three things are noteworthy. First, 40.Z kg/ha for row planting. and 70.8
File: AAWS49i/Mah.2 7.32 Date 17/10/1984

kg/ha for double ploughing were the best responses observed among five replicates giving grain yield.. Three planted replicates failed totally and five prepared plots were nopt planted. Multiple risks faced by traditional arable farmers would seem to make costly inputs that only increase yield potentials when conditions are good ill-advised.
TREATMENT -----------------------------------------EARLY PLOUGH + EARLY PLOUGH e
20 Kg F/ha 287.4 479.3
Control 247.2 408.5
Response 40:" 70.8
Second, even at this yield response level the return to investment was only
8.3 kg grain per 50 kg bag of single superphosphate (8.3 percent) for row planting and 14.7 kg per bag for double ploughing.
Third, farmers still displayed an :nterest in fertilisers. This commodity may certainly fit more easily into the traditional system than do changes in tillage-pianting.
Nitrogen (LAN) was not applied as a side dressing as proposed. The drought conditions of the season made N applications unadvisable. COWPEA AND MIXED CROFFI-G COMPARISONS
Three researcher managed and farmer implemented (RM,FI) tests were designed. The first of these involved a comparison of cowpea varieties recommended by DAFS plus an early maturing cowpea/sorghum miziture. The second was a comparison of cowpeali'ghum mixtures with farmer levels (3.,000 plants/ha) and research station levels (30,000 plants/ha) of cowpea
-in the mixture. The third involved seeding mixtures of cereal
Thirty farmers were contacted and agreed to put. in one or more test Each farmer agreeing to a test was provided with seed, divided into the five lots comprising test treatments. About half a heatare (approximately 4000 sq.m) was pegged and divided into iive equal sub-plots. Farmers were instructed to broadcast the appropriate seed -- indicated on bags numbered I 5 -- and plough it it as soon as possible following the next planting rain.
Many replications of each test arranged for in Shoshong East were never implemented. This was due to the few planting rains during a period extending from the end of November through January. It also teflects the
File: .AAWS49/Mah.2 7.33 Date: 17/i1011984

difficulty of getting farmers to implement trials, even when provided seed,
when they are concerned about minimal acreage of their staple grain.
Several replications of all tests were implemented in both Makwate and Shoshong East but all replications of tests two ;nd three failed in the
drought. Seven replications of the ccwpea variety comparison established
measurable stands in Makwate. Of these, five produced grain yields for
varietal comparisons. Table 7.17 gives these varietal comparisons based on
five observations and seven observations for grain yield and stand
establishment, respectively.
"Tswana Bean" 143.9 a 13.68 b
Black-eye 14t.4 a 17.08 a
ER-7 27.2 b 8.65 c
Means in the same column followed by at least one common letter are not dIfferent at a 0.05 level of significance.
In this comparison, both Blackeye and Tswana cowpea from the Seed Multiplication Unit at Sebele emerged .better than the .smaller seeded ER-7 in these traditionally broadcastiploughed plots. Elackeye emergence was also significantly better than Tswana. cowpea emergence, The emergence advantage of Bla'ckeye and Tswanta cowpea was most obvious when planting moisture dnditions were poor. The nature of this advantage is not determined. It could be related to seed size which would be determined by both variety and conditions under which the sseds were produced. Physiological traits related to variety may also be influencing emergence under these conditions. It is possible that SMU cowpea varieties differ significantly in their appropriateness for different tillage-pianting systems or plInting conditions.
Tswana and Blackeye also produced significantly higher grain yields (Table
7.17) than did ER-7. This was do in large part to emergence differences. The strongly indeterminate and Late maturing Tswana cowpea was able to utilise late season rains and give compensatory growth for gaps in the stand. Blackeye did so to a lesser degree. Early maturing ER-7 plants that were established generally gave good pod set.
Farmers also indicated some preference for the Tswana cowpea as & source of spinach over both BWackeye and ER-7,
Because of below average rainfall in both village areas, weed control was generally not critical. Controlled weeding studies were not thought to be
File: .AAW849/Mah.2 7.34 Date. 17/10/1984

relevant under these circumstances. The idea of using a single r ow cultivator for weeding row planted trial plots was also abandoned because the irregularity of the farmer implemented rows made passes wi th the cultivator nearly impossible. LANDS AREA VEGETABLE PLOT
This was ai design/demonsttationt scheme f'6usad on intensified production of green'i ealies, beans .and "ccurbits.. The concept is thought to be most relevant for farmers with a strong incInation toward permanent lands area settlement. The proposal included diligent replanting gap filling early ploughed acre at a high moisture potential site in the field. Manure was to be applied at 15 tons/hectare.
One site in one field in Shoshong was selected for this scheme but the plot was not implemented. The farmer has a strong settlement position in the lands area including a well Because his draught team was in poor condition, the farmer hired tractor for his early tillage and decided not to implement this design trial that required intensified inputs.
Because seed placement is a potential issue with the low input single combined tillagelplanting system currently practiced, ATIP continued to test the government subsidised Sebele poicugh-planter. The tractor drawn two furrow plough unit and the animal drawnwsinglie furrow plough unit were used by seven and eight ATIF farmers, respectively, in 1982-83. For 1983-84, the test of the units was left as a farmer managed, farmer implemented (FMI,FD) trial for those who were interested.
No interest was shown in either village in the single furrow plough unit. Several farmers did express interest in the tractor drawn plough unit in Shoshong though the unit was never used. The unit for an animal drawn two furrow (P23) plough was taken to Makwate and showed considerable promise The attachment bracket for the unit received was incorrect for the farmer's plough. Several modifications were made in iMahalapye and the unit worked quite will. It was still noted that at donkey speeds, the throw of the furrow slice will only cover the dropped seed it the seed box is precisely positioned. Unfortunately, plantings with the unit failed in the drought.
Testing the plough-planter should probably not have been left to FM,F trials at this stage. RM1,RI or RM,F7 trials are still needed to assess the requirements of the technology.
ATIP observed in 1982-83 that seed provided by the Seed Multiplication Unit at Sebele often gave a superior field emergence vis-a-vis the farmers own harvested seed. it was felt that this could b: dud 1to seed treatments unique to the government supplied seed. The planting of treated seed is one of the current DAFS recommendations A farmer managed and implemented (FM,FI) study was set up to assess the benefit for emergence of treating the farmer's ow'n harvested seed with daoptan fungicide.
File: .AAWS49/1ah.2 .- 73 Date: l0/1O/1984

Fifteen farmer seed lots were treated in Shoshong and one in Makwate.
Locally harvested seed was generally not available in Makwate because of
the total drought failure of the previous season. treated seed lot
was matched with an equal portion of untreated seed. Farmers were
instructed to plant the two lots in equal and adjacent plots. Twelve comparisons were planted A11 planted comparisons went in early and
A substantial amount of data were collected during year one through
technical monitoring, trials and. surveys. While data were subjected to preliminary analysis during year one, a goal of year two was to conduct
further data anaJysis in order to prepare reports to circulate to GOB
officers in DAR, DAFS, and DPS, and to other farming systems projects
active in Botswana.
Report preparation was postponed until after the activity of the cropping season, in light of the opportunity costs of time spent on data analysis.
Moreover, trips to Gaborone to analyse data and prepare reports proved to be difficult, Report preparation began in late MIay when the bulk of field
activities were completed and the Apple Hle became available in Mahalapye.
Four reports baed-on data collected in the first season and two based on
data collected in the second season-are now available as ATIP Progress
Reports (see Section 7.:3). Severi: additional reports will become available
in draft form in:-thenear future.:
A farmer workshop was held in Makwate on 18 May. The purpose of the
workshop was to :review with farmers and representatives of DAFS some key observations stemming from the ATIP trials programme in 1983784. Al: ATIP farmers from Makwate were invited to participate in the workshop, plus ten
S ATIP farmers-from Shoshong East; The RAO Serowe, the Central Region Crop
Production Officer, and the DAO for Mahalapy e East all were invited and
attended the workshop.
The focal points of the workshop. were field visits and a follow-up
discussion session. Attention was directed to two k.ey tria:s: sole plough
and planting methods comparison and cowpea cropping comparison. The
unifying theme was exploration of alternatives to traditional broadcast and
single ploughing of the same crop mixture across a farmer's entire field.
ATIP officers were interested in communicating the message that a strategy
involving multiple tillage and plantng alternatives including the
traditional approach), depending on soil moistu~ e conditions, might be used ::as an improvement over a singular strategy of broadcast and single plough.
It was not clear by the end of the workshop 'that the intended message had
been communicated to participating farmers. Farmers showed only modest
interest in the tillage and planting methods trial. More Interest was
shown in the effects of application of F. In follow-up discussions,
farmers said they see benefits in modified tillage-planting systems but
most were concerned about resource constraints.
A report on workshop activities and farmers comments has been prepared and
File: .AAWS49IMah.2 ? .36 Date: 17110/1984

circulated to attending DAFS and ATI? officers tATIP Progress Report M84-). Copies of the report are available though ATIF Mahalapye.
During the 1984-85 season, ATIP Mahalapye is planning a wide range of activities covering all FSAR stages. Whole farm studies will continue, with a modified format to improve the efficiency with which data are collected. Whole farm economic analysis will play a bigger role than in the first two seasons. Technical monitoring also will continue. Monitoring data will be supplemented with at least one special technical investigation Farmer decision analysis and livestock practices are new areas of descriptive/diagnostic research planned for the 1984-85 season.
Technical trials will involve a modification of previously attempted trials and some new areas of investigation. Resources will be enhanced at Makwate and Mahalapye will be added as a site to accommodate a more efficient RM,RI programme. Shoshong will remain as a predominately RtM,FI and FM1,FI research village and Makoro will be added as a predominantly FM,FI testing site. Several "non-leverage interventions" will be subject to design stage trials. A modified version of the "steps in technology" trial attempted by ATIP Francistown-during the 19832-841 season will be another design stage trial.
Tillage-planting systems investigations will be the focus of testing stage research. An integratedimage-pianting scheme will be tested in RM,FI trials .as ,will alternative planting methods. Techtnical and economic analysis is planned for data collected in the coming and past seasons. Several .indigenous. practices used by ,:some but not all farmers ill be tested andsubject to both techncal and economic screening:
Dissemination stage .activities are being planned for the first time. A simplified version of the tiliage-planting scheme will be the primary dissemination stage activity.
An increasing.proportion of time will be devoted to data analysis, synthesis and report preparation in the neoi t: and future seasons.
Following a pattern established in the 1983-84 season, ATIP Mahaapye will give high priority to maintaining close interdisciplinary. cooperation. Cooperation is reflected in the participation of both technical and social scientists in designing, implementing and analysis of-surveys and trials, and, shared responsibility for implementing some testing and dissemination stage activities.
The primary research activities planned for the 1984-85 season are:
(a). Description and Diagnostic Stage:
Whole Farm Studies
Crop Growth Response .Studies
Weed Control Studies
Regional Perspective Studies:
Farmer Decison Analysis Studies
File: .AAWSB0/Mah 3 7.37 Date: i7/10184

(b). Design Stage.:ehoog ra
Commercial StepsinTcolg ra
Ridge Ploughing Tra.
Non-Leverage interventions "Irials
(c). Testing:
Improved Tillage-Flinting Sy stemsr Inv~est iga t ions
Crop See.dtng Comparisons
Indigenous Practices Investigations
(d). Dissemination:
Tillage -P larting S cheme C p.nstration
Linkages-with Sebele Researchers.
Farmer Workshops
A summary of the proposed research programme is presented in Table 4.
(1I) Multi~le Visit Resource Use Survey
The objectives,. Justilicatlon and approach to. the MVRU have been significantly reformulated f~ar the. .1983-.84 season.a
Objectives For the coming sea-so.In, the primary objective of the DIVRU survey will be to coll ct~ an accurate set o f d ata on arable production enterprises and beer Prewing.. The survey w4ill be -administered. to a small sub-set of the maost successful.,farmers partfcipat ing in the MVRtJ in the pastin. order to obtain, a "best case" a-zaiysis of resource use in arable production and returns to. those resources. The survey,-aliso will b fe administered to a few "c:.miercial- beer brewing househo ds to rvd comparison between commer-ci beer brewing~ and zarable production for households With 'limited resources for arabic production.
A secondary objective of the MVRU will be to comrpatre arable resource. use patterns and returns to resources in the 19P,4-85 season with the two, prior seasons for those' households which managed to have some arable production during the 1982-8SAand 1983-PA drought years.
Just if icat ion :JuIstification for a shift-in-the format of the M!rkU rests on three premises:
(a),. IMany. objectives of "'-te MIVRtJas a. tuy Y hole farm res )irce flows
have been &ccozrpls j~ed, or .can be, accomplished with- a less
frequent, less detalle -d survey format. For example,, household maintenance labour wi no longer be collectedl.since:, after, two
years, We Ihave sufficie-n dati. to understand seasonal. patterns of
labour use for these &ctuivities. AlsS'; we h~eidentified the
primary sources aind uses f cash. Cash flow p akterns do not differ
significantly among households: 'Miost cash-is used for food or
other household goods --and comes from, ocassional remittances or cattle sales. We know fram prio r surveys- that any investment in
File: AAWSzIG!Mah.:.,- 7-38 .7 Date: 17110184

implements will be an infrequent expenditure and that tractor hire
is a major, risky investment.
(b). Fewer ATIP Mahalapye resources should be spent supervising the
MVRU than has been necessary with the prior format. As this is the
third year of the project, testing and dissemination activities
(including technology monitoring) need to be allocated more
resources than in prior seasons.
(c). It has been difficult to gather complete, accurate enterprise data
with the prior format These data are necessary to evaluate
whether cropping enterprises can be competitive with alternative
activities available to households. By concentrating on households
which have the resoucces necessary to be successful arable
producers and which have demonstrated a keen interest in arable activities, we can better be assured of having completed data for assessing the contribution better traditional farmers can make to
national production and employment obectives.
AProach A simplified version of the .MVRU sheets relating to household income activities (former Sheets I to 7) will be used:
(a). The new Sheet i will record household and non-household labour use
on household lands
(b). The new Sheet 2 will record household and non-household labour use
on non-household lands and on non-croppilng activities.
(c). The new Sheet 3 will record data on traction use, whether household
or non-h.ousehold traction, used on either fieldwork or transport.
(d). The new Sheet 4 will be used to record inputs and outputs for all
income activities, arable and non-arable.
Data on non-houseold labour for non-cropping activities, inputs for non-cropping activites, atand non-household traction for transport are being collected for the first time. These data will enable more complete enterprise analysis.
In the revised forms, plot monitoring data will not be duplicated. Plot monitoring data will still be collected by enumerators but will be recorded only on technical plot monitoring forms (see"Section 7 T2.-.2 beloww.
Only 12 h ouseholds will participate in the MVIRU dating the 1984-85 season, 7 from Shoshong East and &.:from t4akwate. All 12 will be households which planted in 1982-83 and 1983-84 seasons. Eight households will be selected from among the most successful households in prior seasons. These households will also participate in technical plot monitoring activities (see Section 7.2. 1.2 below), to reduce time spent supervising data collection. Four additional households will be selected to ensure beer brewers are adequately represented and to enable a analysis of inter-seasonal differences in arable resource flows by households with more limited arable resources.
(2). Economic Analysis of MVRU Data
File: .AAWSSO/Mah.3 7.39 Date: 17/10/84

Objectives : An important office activity during 1984-85 will be to analyse data from the first two seasons. The primary objective will be to prepare and circulate at least one ma or report on whole farm income and resource use.
Justification :Sotswana has an abundance of agricultural. statistics relative to most African countries. The Farm Management and Agricultural Statistics Units of the Division of Planning and. Statistics collect farm management data each year. However, the information derived from these surveys is largely descriptive and generally is presented for spatial groupings of households.
The primary purpose of the proposed data analysis is to diagnose farmers' problems, sources and uses of resources, and the productivity of alternative technologies. Analysis will be carried out using groupings of households based on human endogenous elements of farming systems. Circulation of results to DAFS and DAR personnel should help them, as well as ATIP officers, evaluate the relevance of alternative technologies and extension activities for particular types of households.
Approach : Data analysis will concentrate on whole farm income, enterprise budgets, .labour use patterns, cash flow analysis, resource use regressions, and whole farm production functions. MVRU data will be supplemented with trial and plot monitoring data. Diagnosis will stress relationships between arable activities and other household activities.
(3). Activity Survey
Objectives : The objective of the Activity Survey is to provide a qualitative profile of household and non-household labour use for household income generation, traction use, revenues and expenditures, food consumption, and animal I venitory changes. Analysis of the profiles should increase .understanding Tof the relative importance of arable,. livestock and non-agricultural enterprises during the coming season for households with different endogenous circumstances
Justificat ion Through the MVRU 'usrvey, ATIF Mahalapye collected data on whole farm resource flows for 27 households during the 1982-83-and 1983-84 seasons. The small sample size limited the number of cases for evaluating household activity for houthoids with particular characteristics, Also, the 1982 83: and 1983-84 seasons were drought years and, therefore, the overall pattern of household activity was likely to have been atypical.
Additional ases are necessary to evaluate the relative importance of household activities, for households with different characteristics and in seasons with different rainfall patterns. Information on household activities is used to gain a systems Dersp ctivre, 6or prescreening improved practices for directing technology development research.
Approach The Activity Survey will attempt to gain a qualitative understanding of the activities of a larger sample of households than has been surveyed under the prior MVRU format, but with a minimal investment of research resources. Approximately forty households will participate in the Activity Survey; those ATIP cooperating households which are not
File: AAWS51Mah.3 7.40 Date: 17/10/84

participating in the IVRU but which will be subject to whole field monitoring (see Section 7.2.1-1(6)) Each h-ousehold will be interviewed once a month.
The Activity Survey will have 5 sheets:
(a). Sheet A: Labour Activity. Each potentially active household member
will be contacted to ascertain his/her level of participation in.
household fielcdwork, livestock and non-cropping activities.
Participation.will be recorded as: everyday, .3-4 times a week, 1-2 times a week, infrequently. during the month, or not at all. The
household head will be asked to indicate the number of
non-household men, women. boys or gicls who have participated in
household income activities. For each non-household participant, a
qualitative level of participation will be indicated, as for
household members
(b). Sheet E. Traction Use The same frequency code from Sheet A wil
be used to indicate use of household and non-household traction for
fieldwork and transport if non-household traction is used, means
of access to that.traction will be. recorded.
(a). Sheet C: Revenues and Expenditures. The frequency codes will be
used for. the different categories of revenues and expenditures
formerly appearing on the former. MVRU Sheets 9 to 12. Categories
of revenue levels from sales or remittances and for equipment expenditures also will be recorded -(eg, ..F 1-IS, 16-30, 31-60,
(d). Sheet D: Food Consumption. The frequency .codes will be used to
indicate frequency of consumption of: milk, beef, goat meat or
mutton, wild morogo. field morogo, melons, cowpeas: sorghum bogobe,
and maize bogobe.. i addition an approximate estimate will be
recorded for total household consumption of sorghtum and maize meal
during the month(e). Sheet E: Livestock inventory Changes. All changes of livestock
inventory will be recorded as they were on the former MVRU Sheet
13. In addition, vaccinactiotis and. dippings will Ibd recorded.
Monthly profiles will be compared for households grouped according to resource endowments, demographic compositions, traction recommendation domains, and arable activity.
Sheets C to E also will be administered to households participating in the MVRU during the last interview of each month.
(4). Inventory Survevs
Objectives : The objective of household inventory surveys is to identify the demographic composition of households: and the pattern of household assets, And then analyse relationships between household endogenous circumstances and arable activities and cropping outcomes. This is a continuation of an acti ity initiated in the 1983-84 season.
File: .AAWS50/Ma.h.3 7.41 Date: 17/.0/84

Justification A71P Mkhalapye research to date indicates differences in arable production problems and the relevance cf alternative solutions are strongly related toa erid:genous circum stances at household ds.. Therefore, technology prescreening and targeting requires information on relationships between endogenous circumstances arnd outcomes,. and howt these vary across seasons.
Inventory data is also needed to computer% returns to the livestock enterprise.
Approach I The same inve ntory firms will be used in 1.984-85 -as were used in 1983-84. This year, inventory form s will be &dininstered to all households coo perating with A7171F Mahalapye. increasing the sample sire to 5t households. The' livestock inventory will be adirinistered'twice, in October, 1984 and April, 1 985. The Hiousehold -Cansus will be administered in October, 1984. The Farm Fired Capital will be administered in Februairy-Marth, 1985.
(5). Whole Field- ?'onilo ring<
Objectives The objective of whole field monitoring is totain data for the analysis of reDlat-_ornshzrs 'bttwean farm resources, enIvironmental parameters uid whole farm. crop arod u ct Ion. components suih -as days of operationss. rates per day, :total a re a cultivated, crc~p prol ecl Lon and total crop production.
Justification I n the two ppreviou5 seasons, field level moni*t-oring of crop production was based on Aggregation of plot level data. This has provided gpod data but on a Limited number of whole field 6bservations. This limitation is particularly impc-rtant in analyses that partition the data inito farm or field type el~legories, -in !?t4-85, a simplie -r whole farm assessment that provides suffrrricnt data for whole farm analyses Will-be carried out on ain eipanded number of farms.
Approach :Whole field..tec:hnicai m onitoring of in'puts and crcopping outcomes wr411 consist 'of wihole> field assessments 6n two visits The first visit will. be at the completion:*of ploughingiplanting and the, seaCon d wjill be post-.harve-st. Assessments will be. as follotws.
(a). First ,isit:Area-.s cultivated by secitlon lsectionr is definedJ as the toal area
planted to a given c.rop or crop mixture on s; single: planting
moisture period'
Days of ope rations
Cronp/variety selections
Whole farm, seed requirements
Qualitative assessment of stand establishment by section
(b). Second visit:
Rating, of 'cropping outcome by. se-Iticn
-Whole farma icop protection acivites.
Whole -farm crop production IjV ccopivarieti
Determinants in thle angalysis include lie nousehoid resource situat on an environmental factors such as field conditions and climatic pattern for that
rile: .AAWS50/14ah.3 -7A42 -Date. 17/10/84

farm. Pertinent household characteristics will be assessed on all new farms in the study. Soil type and fertility have been assessed for most ATIF fields. This data set will be completed in 1984-85 Rainfall patterns have been assessed for 33 sites in the research areas and this will continue. Rainfall patterns for new farms will usually be determined from data collected at the nearest existing tain gauge. Additionally, at least one technical survey will be conducted to support the whole field study assessing parameters such as border vegetation, topography detail, etc. .This survey will be handled by the Mahalapye technical staff in cooperation with the village T4 extension staff.
(6). Livestock Practices Survey
Objectives The primary objective is to assess differences in cropping practices depending on herd sise, livestock management practices and settlement pattern of a household. A secondary objective is to evaluate whether "progressive" management practices in livestock enterprises are complementary or competitive with "progressive" crop management practices.
Justification : ATIF is mandated to concentrate on arable production. However, livestock sales are the most important. source of income for most rural households. Cattle and donkey traction are used by the vast majority of crop farmers, Milk from cows and goats are a vital dietary supplement< throughout the summer, not to mention the above average consumption: of.:meat by Batswana. In brief, no one would deny the!key role of livestock in Botwana.
From a system perspective_. a mandate to concentrate on arable production cannot be interpreted as a justification for neglk ting household livestock activities. Nearly any change in arable practites"is, from a farmer's perspective, interpreted in relation .to impact .on" iviestock enterprises.
Information on livestock practices also is : needed by the ATIP animal scientists in order to evaluate alternatives:for increasing productivity of household livestock enterprises.
Approach This study will entail two ,survey activities, designed in cooperation with. the ATIP animal scientists. First, a single-visit survey on livestock management practices, livestock enterprise productivity, and system linkages -between livestock and arable activities will be administered to ATIP Mahalapye farmers. Second, a series of exploratory visits will be made to cattle costs of selected ATIP Mahalapye farmers. As with FSR exploratory surveys, the focus wil be on management practices, problems and priorities from the perspective of those actually engaged in: large stock management. CHOPIGROWTH RESPONSE STUDIES
(1). Plot Monitoring
Objectives : The objective of plot monitoring is to ascertain crop responses to farmer and environmental inputs. Analysis of these data is intended to provided a probabilities estimate for conditions that make the traditional low input tillage/pianting system a good bet. A third season of plot monitoring would also contribute to improved description stage
File: .AAWS50/Mah 3 7 43 Date: 17/10/84

analysis if there were to be a significant seasonal effect on farmer inputs.
Justification : Third season observations will be used to test the hypotheses generated in the first two years. Additionally, new variation in crop responses would be expected in the event of a measureably different rainfall pattern in the third research season.
Approach : Plot monitoring will be continued on fewer farms than in the first two seasons. All plots on only five farms in Shoshong, five in Makwate and two in Makoro will be monitored. However, only farms that have had a large number of planting days in the first two seasons of the project will be selected. This approach will provide a substantial number of observations with a l-gistically efficient number of sites.
Little change will be made in the nature of the plot data being collected and the manner in which these are obtained. The technical summary. sheets described earlier (see Section 7. 1.i.1 2) have proved useful and will be modified only slightly for use this season. ..For .198q-85, technical plot data will be collected by the village based enumerators.
Because soil moisture zs such a s gnificant factor .n crop stand establishment within the traditional system,.soil moisture assessments at tillage/planting are critical. Misture assessment will focus on the furrow slice because this is the sit"eof seedling establishment. A methodology study will suppor rt nitoring of. soil moisture during stand establishment. Three criteria for evaluating soil moisture at planting will be considered by ATIP Mahalapye. These are considered because of their potential relevancy in recommendations to farmers. The first is being developed from, soil texture and day air temperatures. 'The seond is based on precise local (Setswana) characterisations of moisture conditions. The third is a soil moisture estimate developed from the FAO model of soil moisture based on evapo-transpiration potenial. Any of these criteria that are employed will be verified through gr.avimetric: or other type analysis of actual soil moisture in the critical so:il zone. Verification sampling will be handled by the T4 in the expansive Shoshong "Lands" area.
Monitoring of soil moisture for a greater portion of the soil profile and through the entire growth period might help explain variations in crop development. This would be particularly useful, in the comparison of tillage treatment effects on infiltration of rainfall on.farmaer's fields. Because it does not necessarily provide pertinent data for stand establish nt, the costs/benefits of full season profile monitoring need further evaluation : "
(2). Weed Control investigations
Objectives : There are two objectives to this study
(a). To further determine the pri-ority of weed control research by the
ATIP project. This consists of both a quantified description of
the range of weed problems encountered and diagnostic assessment
of farmer and enironmentailinputs that contribute to weed control
within the currently practiced..(ow weeding input system.
File: .AAWSS0/Mah.3 7.44 Date: 1710/84

(b). To feed information on the observed traditional pattern of moderate
weed burdens up to the newly established post of Weed Agronomist at
The first is a descriptionidiagnostic stage activity. The second is part of a general interest in improving linkages with component researchers at Sebele. A third weed control investigation at the Testing Stage is described in Section .
Justification A wide range of weed levels have been observed on fields and on plots within fields, in the project areas. Plots that are nearly weed free without weeding can be found among weed problem areas. Weed problems can be attributed to one of two sources: first, poor burial of grass and other weeds at ploughing. and second, fresh emergence of perennial plants and annual seeds. It is hypothesised that the first type of problem is related to weed levels at ploughing, soil moisture at ploughing, ploughing depth, ploughing speed -- a function of traction type and animal number 'The second type of problem should be related to ploughing depth and post-plant rainfall. T.his combination of management and environmental factors could have relevance for improved weed control recommendations. Traditional management factors or modifications of these factors that contribute to weed control may find quicker acceptance than would recommendations that require more hours weeding.
Approach : This investigation includes description/diagnostic monitoring activities. An evaluation of the current range of weed problems will be continued using the weed growth survey initiated in 1983-84. This survey is conducted by ATIF technical staff. The survey assesses incidence, stage of growth and percent area covered for important weed species in the Central Agricultural Region. .n the assessment, comparisons are made between different cropping patterns, rainfall patterns and periods of the season. Comparisons are also made between the effects of various experimental tillageiplant ing systems on levels of weed control. The Sebele Weed Agronomist will contribute to the design and analysis of this survey.
It was intially planned that regional perspective studies would begin during the 1983.-84 season. The general objectives and justification for regional perspective research were presented in ATIP Annual Report No 1.
(1). Regional Ex oloratorv Survey
Objectives : The primacy objectives of a regional exploratory survey are.
(a). To identify where in the Central Agricultural Region changes in
farming systems seem to be taking place or have taken place in the
last 10-20 years.
(b). To find out why donkey and tractor traction have been replacing
cattle traction in some villages but not in other villages.
(c). To rank arable production problems and research priorities based on
File: .AAWSS0/Mah.3 7.45 Date. 17/10/84

farmers' perceptions, and relate rankings to village types 'eg,
size, access to large villages, primary traction type).
d). To list perceptions of advantages and constraints on adoption of
technologies currently being recommended by DAFS or tested by ATIP
Justification : ATIP Mahalapye needs to:
(a). Continue to identify location specific priorities throughout the
Central Agricultural Region
(b). To take advantage of the diversity of circumstances found in the
Region to identify solutions and evaluate how proposed solutions
might be incorporated into farming systems.
Approach : District Agricultural Officers (DAOs) and ADs will be contacted first and characteristics f tensionn areas reviewed. Then several extension, areas in each district will be visited and farmers contacted. Throughout the survey, a range of hypotheses on farming system problems, trends, attd priority technical changes will be discussed.
(2). Quarterly Price Survey
Objectives The objective is to gather information on trends in prices of agricultural inputs and commodities primarilyl y grain and meal) from a selection of villages in the Central Agricultural Region.
Justification : There are two key reasons for research on prices of agricultural inputs and commodi ties:
(a): Prices are the best indic.itors of the value of agricultural
commodities and the cost of resource required to produce those
commodities but there is little information available on .price
trends in different markets (informal markets in lands areas and villages, formal markets in small villages, and formal markets in
large villages)
(b). Many farmers appear to invest in arable production to produce
household consumption, rather than to produce a surplus to sell.
This strategy is necessary for food security if markets are not
integrated. If markets are well integrated, households can devote
resources to activities which maximise income and then purchase
food. Trends in prices in different markets can be used to
evaluate whether markets are integrated.
Approach : One to two small villages will be selected within 40-50 kms. of Mahalapye, Serowe and Palapye. in each of the large villages and near-by small villages, a sample of local trading stores will be identified. Also, a small sample of. village and lands area dwelling units will be identified in each village. The training stores will be contacted once every three months to find out prices of arable inputs and ma3or agricultural commodities. The selected dwelling units will be interviewed to elicit information on exchange prices for foodstuffs in informal markets."
(3). ALDEP Particinant Survey
File: AAWSSO/Mah.3 7.46 Date: 17/10184

Objectives : The objectives will be to identity problems ALDEF participants have encountered in obtaining loans (now grants and to assess perceptions of the impacts of different loan packages on household productivity. This information will be made available to the Regional ALDEP Officer.
Justification : ALDEP probably is the most important agricultural development programrame for resource poor farmers. Now that the programme entails substantial grants to farmers, it is important that priority is given to packages which have the greatest impact on farmer welfare and productivity. Also, since substantial numbers of farmers ate likely to apply for grants, it is vital applications can be efficiently and equitably processed.
Approach : A list of pending, approved and rejected ALDEP applicants will be obtained from the regional ALDEP officer,. Emphasis will be given to current participants in the ALDEP programme. Approximately 10-15 ALDEP applicants from each of 6-8 vi.ilages will be located and interviewed over the course of the 1984-85 season. A similar study is planned by IFFP and results will be compared. FARMER DECISION ANALYSIS STUDIES
Any change in farming practices must begin with a farmer s decision to make such a change. The farming systems approach postulates that farmers' decisions are influenced by the element and the exogenous and endogenous human elements of farming systems: Two component decision studies are planned to pro vide information on circumstances influencing decision outcomes:..
(a). Decision Unit and Management information S-tudy.
(b). Farmer Terminology Study
(1). Decision Unit/Managem.ent Information (DUM.D S)tudy
Objectives : This study has two closely related objectives.
(a) ,. To clarify patterns of decision making among individuals, and at the
household and supra-househdlid levei, over time.
(b). To identify information -- including subjective information and
information about informal institutional arrangements -- used by
relevant, decision m ikers in making key decisions regarding
practices and cropping outcomes.
Justification : There is ample evidence from t roughout Africa which suggests. decisions affecting farming practices and outcomes are made at individual, supra-household and village levels -- as well as at the agricultural household level. It is important to understand which decisions are made at the individual or household level, and which are strongly influenced by supra-household institutional.arrangements. This will facilitate prescreening potential techno logies for their impact on particular members of househrtIds and on networks 'of households
Approach :.The study. will entail informaIl, exploratory discussions with
File: .AAWS 0/Mah.3 7.47 Date: 17/10/84

ATIP farmers, followed by a single-visit survey. Through exploratory discussions a preliminary listing will be made of:
(a). Key arable decisions and who is responsible for which decision.
(b). Information about technical. and human elements of farming systems,
particularly institutional arrangements used by decision-makers.
Based on exploratory contacts, a verification survey will be designed and administered to a random sample of farmers. The sample of farmers will be drawn from ATIP research villages A similar study will be carried out by ATIP Francistown to enable a comparison of results.
(2). Farmer TerminologyV Study
Objectives : The objective is to develop a listing of key terms used by farmers to describe their techical and human environment. Farmer terminology can be used to identify distinctions which are important to farmers and knowledge of popular terminology can facilitate dissemination stage activities.
Justification : Recommendations must be based on distinctions farmers recognise and can understand. Researchers are likely to be more successful if they trV to build on existing distinctions rather than trying to get farmers to relate to distinctions the researchers uses,
ADproich : Continuing an activity started during the 1983-84 season, farmers and GOB agricultural officers will be interviewed inf6fmally to identify terms used for soils, topography, weeds, soil moisture conditions, and various institutional arrangements. Lists will be compiled for several areas in order to compare similarities and differences in terms.
A basic tenet of the FSAR is that households can be divided into relatively homogeneous groups for research aind extension. AiP Mahalapye primarily has grouped households using a Rb scheme derived soon after research began. The recommendation domains (RDs) are based on type of traction used and access to traction.
During the first two years, it has been found that households in the same traction RD have different resource endowments and orientations toward crop farming. Moreover, households have shifted across domains with much greater rapidity than was expected. It has become apparent that a classification of households taking into account resource endowments and managerial orientations might be useful for directing research activities
Tentative farmer classifications have been developed through two years of interaction with ATIP farmers. The classification scheme, .based on resource endowments and arable activity, as well as researcher subjectiveassessment, is presented below.
(a). Code 1. Above average resource endownimerinticommercial practices [A
configuration of DAFS recommendations, family may have a commercial
outlet for surplus produce.
File: AAWSSO/Mah.4 7.48 Date. 15/10/84

(b). Code 2: Average or better resource endowment/progressive traditional practices. CProgressive traditional implies diligent
and thorough operations; practices such as double ploughing to
control weeds, replanting, early weeding are common.]
(c). Code 3: Good resource endowment/solid traditional practices.
ESolid traditional practices are specified for two traction groups.
First: with animal draft, an unspecified number of days of
operation but certainly exceeding 4, broadcast/single ploughing of
croplvariety mixes, ploughing depth between 5 -15cm averaging 10cm,
no use of fertilisers, row planting, early or double ploughing.
Second: with tractor hire; as above except that the number of days of operations is reduced but with a comparable hectarage Farmer
intention is still to include more than one day of planting.1
(d). Code 4: Medium or weak resource endowment/solid traditional practices.
(e). Code 5: Weak resource endowment/struggling traditional practices.
[Struggling traditional indicates similar practices to solid
traditional but without the hectarage of the solid traditional.]
(f). Code 6: Weak interest regardless of resource endowment/arable
activity less than standard traditional and much less than
potential for the resource endowment
The above division of farmers will be used to identify target farmers for technical trials to be conducted during the 1984-85 season. COMMERCIAL STEPS IN TECHNOLOGY TRIAL
Objectives : The purpose is to assess the technical and commercial advisability of various inputs including current DAFS recommendations This is a technical package design trial targeted for farmers willing and able to substantially increase their arable production inputs (Codes 1 and
Justification A small .prop rtrit. of: fatimrs ith the Central Agricultural Region appear orientated towaids :arable pOduction for commercial surplus These farms may represent} however, a significant potential for increase of grain production in the region. The benefits of such surplusjpoducers has been most apparent these past two seasons of drought. Not only is some quantity of staple produced but possibly of more importance 'ahi ben the multiplication o seed. This multiplication can provide de seed not available in quantity nor .of the preferred varieties from the ,Se'd Multiplication Unit at Sebele.
Primary analysis should be for the financial advisability of inputs for commercial production within this environment. A modified traditional systems perspective is necessary, though, because of traditional opportunity costs (ie, battle herd management) and because village ievel resource ,.pnstraints (ie, availability of inputst labour) still influence arable practices
File: AAWSSOI/Mah.4 749 Date: i315/10/84

Approach : The "steps" study will be researcher managed and impleme dted (RM,RI). Maximising precision on the technical evaluation is considered more important than farmer evaluation. All tillage/planting treatments will be implemented by project staff and staff from the Mahalapye Research Sub-Station. Crop protection, including weeding, is .primarily the responsibility of the host farmer Some consideration will be given to helping with keeping stray cattle away from the plots. This could include both planting a large non-test border around the test and setting up an electric fence unit,
Five field sites will include some or all comparisons of this study. Three farm fields have been selected in the immediate Mahalapye area. The three fields are spaced at approximately 2km intervals with the nearest located 6 km from the Research Station. Another field site, planned for Makwate, will not include the sub-soiling comparison. This site is yet to be located. The Research Sub-Station will provide a fourth location for the sub-soiling component of the test.
Tillage operations in the Mahalapye area wil be carried out with the Research Sub-Station MF 265 tractor and research equipment. Equipment will include a soil ripper, a three disc plough (60 cm), two row; planter, and a spike tooth harrow. Tillage operations in Makwate will be with ATIP donkeys and ATIP equipment. This equipm Ot' includes a single furrow plough, a single row planter and a spike tooth harrow. All broadcast applications of seed and fertiliser will be done with a hand operated "cyclone seeder". Segaolane sorghum seed will be planted in all comparisons of the study Fertiliser applied will be single superphosphate (8.3 percent).
The experimental design will be a split-s plit-strip p16t: factorial with two replication blocks per location. Makwate and the Research Sub-Station wii have only some treatments resulting in a split-strip factorial and a ROCBD, respectively, at those locations. Main plots are the sub-soiling comparison at iahaMLapye and the Research Sub-Station. The sub-unit includes early ploughing and planting methods. Feriliiser is stripped across planting methods and randomised within the early ploughing plots.
The experimental treatments will be
(a). Sub-soiling with tractor to. 45cm" (late July
(b). Early ploughing (October) 15-20cm with tractor in Mahalapye,
10-15cm with donkeys in Makwate.
(c). Phosphate.fertiliser will be applied broadcast on all fertilizer
plots before the early ploughing. Application,% will be 30 units
(d). Planting methods will be broadcast/ploughingr row planting and
harrow seedbed preparation plus row planting. Because the planting
methods will be put in on both early and non-early ploughed. plots, row planting treatments will require a same day or late ploughtng.
The resulting combination of early ploughing and planting methods
comparisons includes:
File: AAWS50/Mah.4 7.50 Date: 15/10/814

Same day plough/row. plant.
Same day piough/harrow/row plant.
Early plough/broadcastisecond plough.
Early ploughirow plant.
Early plough/harrow/row plant.
Five field sites are part of this experiment and an effort will be made to generate as many distinct environmental situations. Of the three fields near Mahalapye, two are on soil of a sandy loam/loamy sand type '.and the third is on a sandy clay loam. These soil types are typical of the soils found elsewhere in ATIP Mahalapye research areas. The Makwate site will likely be located on a loamy sand and, because it- is situated 56 km from Mahalapye, is more likely to experience a different rainfall pattern. Also, to increase the likelihood of a range of rainfall patterns in Mahalapye, farm field plantings will be staggered at approximately 2. 5 week intervals or as possible. Makwate and Research Sub-Station plantings should be between 15th November and 15th December. All plantings will be targeted for good moisture situations.
With reference to non-experimental variables, it is not anticipated that plant population will be controlled following the inital seeding. In the event of a general emergence failure at any location, that location will be replanted. Weeding and crop protection will be the responsibility of the host farmer. These activities will be supervised by ATIP staff and, if necessary, supported to protect the test. RIDGE PLOUGHING TRIAL
Objectives :This design study involves technical evaluation of various tillage/planting uses of the ridging plough.
Justification The ridging operation is expected to improve soil water management. Benefits could result from better root zone drainage following heavy rains and iiap;oved water infiltration Results from research at Sebele on tied and un-tied ridge cultivation have shown some promise. Typical ridge ploughing plus planting requires precise operations and it is not certain how local draught teams can best produce an adequate ridge.
Approach : This design stage study will be condu cted ias a researcher managed:and iplemeted (RM, R ) trial in imJate using ATIP donkeys and equipment The pr opposed t eatments include broadcast/ridge ploughing, ridging (untied) pius row and plus hill (hand) planting. and tdge plough replanting of 'failed first plant emergenie. An attempt wil also, be made to construct wide ridges using both the tidging plough and a 'standard single furrow mouldboard.
In addition to technical evaluation of results, the duration of field operations will be monitored. NON-LEVERAGE INTERVENTIONS TRIALS
(1). Draught Managcement
Objectives : To obtain subjective project and farmer assessment of. .
File'. AAWS0/Mah.4 7.51 Date 15/10/84