Multiple cropping in tropical Asia

Material Information

Multiple cropping in tropical Asia
Series Title:
Multiple cropping in tropical Asia
Harwood, Richard,

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:
183709413 ( OCLC )


This item has the following downloads:

Full Text
J0AN 87

JAN 81976


Richard R. Harwood and Edwin C. Prices/

Present and potential multiple cropping patterns are determined by
a wide range of physical and socio-economic conditions against a back-
ground of available: technology. We have therefore chosen to present
multiple cropping in tropical Asia in several parts which describe first
the economic and physical factors which influence multiple cropping, then
the present farming systems within which.multiple cropping is practiced
and their direction of change,-and then.the major multiple croppingJpat-
terns of Asia. Relevant research .needs are briefly.summarized.

n: Tropical.Asia includes the area between the-Himalayan mountains
adnd'" ten degrees south of the equator from'.:
Indonesia, New Guinea, the Philippines, southern Taiwan, the southeastern
portion of the Peoples Republic of-China and,-ainland southeast Asia.
The dominant crops in the region are ,mainly- water supply.
Cropping systems in the drierregionso-of India and ,Pakistan are based on
co6in,, .sorghum, millet and wheat, while the .wetter areas of these s-ame
countries and nearly all of east Asia, having high rainfall,. are'fice based,

while most of the economic, tropical Asia that affect
multiple cropping are changeable .to some. degree, two of the more permanent
conditions can be.identified:- a;.smallholder structure,,:and, allied with
this,, a highly dispersed production that; leads to high:marketing costs.
Other .factors which strongly influence cropping systems., but which are
more variable over time and loc4ae,, are agricultural-wage. rates,: the levels
of human and physical capital in agriculture, and prices of agricultural
products and of cash inputs. These conditions and their relationship to
cropping systems will, in turn,. be.discussed here,

Farm size

..The distribution sizes in.As-ia (Table 1) clearly shows the
prevalence of small farms in that region.: One-third of the.farms in:
Southeast Asia are less than a half hcdtare in size, one-half'of .all.
farms are of less than 1 ha, and three-quarters are of less than 2 ha.

1/ Paper presented at the 1975 meeting of The American Society of
Agronomy, as .part of-the symposium "Multiple Cropping".i Knoxville;.
'Tennesseej August.24-29, 1975. Accepted for.publication-by the ASA.

2/ Agronomist and Economist, respectively, The Internatonal Rice
. Research Institute, Los Bafos, Laguna,.:Philippines.

2 -

The average farm size in Southeast Asia is 1.8 ha, compared to 1.1 ha
in East Asia, 2.4 in South Asia, and about 120 ha in the United States
(Table 2).

Several researchers have found size of farm and number of crops
per year inversely related. In Bangladesh Amed (1965) found that on
farms of less than 2.5 acres the average cropping intensity index was
167, compared to an index of 117 on farms larger than 40 acres (Table 3).
In a study of vegetable farmers in Taiwan, Menegay (1975) also reported
that "farm size decreased as the relative intensity of vegetable pro-
duction increased." But the available research does not clearly say
whether diminishing farm size over time, forced farmers toward intensifi-
cation or, alternatively, whether it was the drive toward intensive agri-
culture that resulted in small farms.

The latter is hypothesized here. More specifically, we suggest
that (1) in a tropical regime annual.production per unit area responds
positively to management input at much higher rates of applications
than in a temperate regime, and (2) the resulting economic incentive
to increase the management input has directly contributed to the reduc-
tion in farm size in Tropical Asia. The argument is given in greater
detail below.

Because tropical climatic conditions are suitable for field crop
production during much of the year, cropping pattern alternatives are
numerous. In fact, as the number of possible sequential crops per year
increases, the number of pattern alternatives expands very rapidly
indeed. At one of the sites in the Philippines where IRRI cropping
systems research has been conducted for about 2 years, more than 170
cropping patterns have been identified among 50 farmers. More than
100 different farm operations are performed on those crops.

Even where lowland conditions limit the number df species that can
be grown, many alternatives in the timing and technique of planting,
and in water control and choice of variety, all multiplied by the number
of sequential crops, result in an extremely high requirement for mange-
ment decisions. The latter situation is demonstrated particularly well
in Bangladesh, where a wide range of rice and jute cropping systems can
be observed within a very short distance along the road. Each system
is minutely tailored to the land and water conditions of a small area,
and to the 'farmer'is other resources and his assessment of the future.

Choosing the most profitable pattern likely requires a higher
level of management where sequential crops can be grown than in one
crop regime. Put another way, marginal return to management remains
positive at'much higher rates of application under tropical regimes than
under 'temperate ones (Fig. 1). Incidentally, this says nothing of the
relative values of the marginal or average product of management under
the' two reginies,. The maximum return across variable management levels
in tropical regimes may either .be lower or higher than in temperate ones,
depending on the quality of other resources..

3 -

_.,The high complementarity between management and physical resources
in.the tropics, suggested here, is a hypothesis that has yet to be tested.
Even more tentative is the..following suggestion of the manner in which
high ratio.s of management. to land have been reached in Southeast Asia.
In .the absence of education which might have augmented the management '
capacity of farmers, and before management-augmenting capital or techno-
logy became available, higher ratios of managerial capacity to land were
achieved through the reduction of farm size. The size of holdings con-
tinued to decrease unti1 the marginal return to farm operators declined
to the level of the labor wage, or whatever was the highest level of
return to managers in alternative employment.

One might further speculate that the same process that resulted
in more managers also produced a high proportion of laborers, as
siblings and offspring of farm operators. -Therefore the redundancy'
of agricultural labor.has:perhaps been a by-product, not the.major cause,
of the drive to achieve intensive resource management. Whatever surpiis
labor that remains in Southeast Asia today may be a condition that is';"
largely incidental to the existence of intensive farming. Of course,'
the rational operator tended to use more labor when it was cheap than
when its price was high.

That cheap labor is not necessary for intensive agriculture is
perhaps indicated by the data from East Asia in (Table 2). Taiwanese -
farms, for example, are now comparatively mechanized, occupying 42 per-
cent of the labor force, compared with 70 percent for Southeast Asia'as
a whole. Yet the cropping intensity in Taiwan is far greater than in
any part of Southeast Asia, about 175 (Chao, 1975). The index levels
in the more labor intensive locales in Southeast Asia do not exceed 150
(Manu, 1975; Oshima, 1973),

Seen in this light, it was perhaps not population pressure that hau
forced agricultural'intensification, .but rather the reverse. The evolu-
tion of small farms in Tropical Asia may have been a rational response to
profitable opportunities for the intensive land cultivation that is
possible in the tropics, Two relatively -p.ermanent features associated
with the long tropical growing season, which .strongly influence crop-
ping system, are high complementarity. of the. management input with land
and a consequently small optimum farm-size ... .


,::Associated with smallholder agriculture, are high marketing costs-,
for. agricultural produce. Numerous.middlemen and transactions are re-.
quired to collect produce from many small producers and distribute it to

Summarizing some of the work presented at a workshop on multiple
cropping, Oshima (1973) concluded that ".... Small amounts of surplus
output from tiny, enumerable farms producing more than one commodity

4- 4

require a wide network of assembly and collecting stations, small pro-
cessing plants, and marketing stalls and centers." This suggests that
the income from the sale of these surpluses are small and scattered
requiring -mall shops and stores to sell farm inputs and consumer goods
to the small farmers. Accordingly, small farms mean small stores, out-
lets, and other selling and service units."

A number of studies have shown markets in Southeast Asia to be
quite efficient, meaning that the difference between the price which
the farmer receives and that paid by consumers closely reflects the
cost to middlemen for performing their functions (Price, 1973; Thodey
and Anee, 1974; Mears, 1974). This feature of high marketing costs has
caused some misunderstanding between economists and agronomists, not -
unknown to those studying cropping systems. Having developed the bio-
logical technology for'producing a crop to fit an ecological niche and
perhaps noting that a quite respectable price for the crop prevails in
central markets, the agronomist is likely to declare that production
problems have been solved, and only economic ones remain. In truth,
it is the structure of production that often causes the unprofitable
operations of smallholders, a problem neither scientist can alter.

Government policies that would encourage the production of similar
crops so that a critical mass of product can be efficiently gathered for
market, are a possibility. The grains of such measures, however, must -
outweigh the losses of the product caused by planting standard, but sub-
optimum, cropping patterns on the unique combinations of resources on
different farms.

To some extent, the high density of Southeast Asian populations
compensates for the marketing problems associated with the smallholder
by providing consuming points that are generally near production.
Southeast Asian produce travels much less distance from producer to
consumer than in many developed countries.

A more changeable component of marketing costs is transportation.
In recent years new roads and highways have linked the interior of most
countries to the population centers. This has increased the cropping
alternatives for outlying areas, as in Northeast Thailand, where a
number of new field crops have entered cropping patterns both before and
after rice, or where the new crops have replaced rice altogether.

Superimposed on the relatively permanent economic features of
Tropical Asian agriculture just discussed, are certain influential con-
ditions that recently have changed as at no time since world War II.
Those conditions which now appear to be rapidly shifting, with strong
implications for cropping systems, are the availability of labor, levels
of human and financial capital in the rural sectors, and product demand
and input prices.


Agricultural labor and wages

As suggested above, the large labor pools of Southeast Asia may
'haV"-~ resultedd from the spurts of entrepreneurial response to the oppor-
tuni'tes -foi tntensifi6ationi. :f-fthere was a causal link between the '.
diiv'i6war~ d intensive land management,, and-growth of the labor force,'--
if'wa's broken when education' and' manageent-augmenting capital and tech-
ti6o1f'VbedameI available, increasirig numbers of decision-makers-per unit
lKid 'ai~a wde ie6'0 longer necessary -and the most explosive'force in po-
pilati'6' gdrwth relented. From then industries gradually ab'sobed'the
"-~idundai labor. In every recent years industrialization has been
-fairl- iapid in Southeast -Asia, and surplus labor has disappeared in
many'irts. Th't real agrictultural:'labor wage is already high in 'falaya;
it appears to be edging up in Thailand, and is likely now to move higher
in the Philippines. On the other hand, the agricultural labor wage is
still quit "low 'in indonie'sia~--'at'B'less than half the rate that is paid
in the Philippines and Thailand.

Human' capital ".:

A burgeoning class o agicultural technicians educated in both the
newly advanced curriculum of Asian Universities and in Universities
abroad, is now of size to have a measurable impact on the amount of tech-
nology available to and used Iby" farers.''As seen in Table 4, the number
of research scientists and th" number of' extension workers in Southeast
Asia each about doubled.between 1965 and 1974. In South Asia the rate
of manpower increase, though lower, still is substantial. Furthermore,
in 1974, South and Southeast Asia-combined had 36 percent of the world--s
extension workers (not shown), but0only 'bout "4'ip"rcent of'the agricil-
tural researchers (Boyce and Evenisbonji,975)'".'" --

Financial capital

The investment of financial capital is also-increasing, assisted by
aggressive government programs which provide low-int'ieest-rate loans for
either operating or investment capital in agriculture. A recent Presi-
dential Decree in the Philippines requires that all banks operating in
the country allocate 25 percent of their Ib'd:i lue to agriculture.

Perhaps the programs have been too :-r'.all,or they havebeen too
restricted as to purpose, for still the ~uiral interest rate is high.
Farmers must 'till roly heavily upon private lenders whose: rates freqbent-
ly exceed 100 percent per- annum. An estimate''of averagee Interest, rate
on loans from all sources in Northeast Thailand-Ij" 80 pe'r6et' per 'annum
(Price: 1973), .

-6 -

Product demand and prices

'Also affecting cropping'systems are the prices of products produced
which, in' turn, result partly-from the level of product demand. .Two
factors have radically changed:the demand for agricultural products in
"Southeast Asia: the entry-of aggressive buyers for the large Japanese
markets, and the depletion of U.S. grain reserves which for 25 years had
helped to stabilize world prices. ;;The former-development has not only
increased the demand for commodities with established markets, but has -
created markets where none, or only weak ones, had before existed. Soy-
bean, corn, and sorghum buying particularly by agents who willingly seek
'out and -contract for small but long-term supplies, has had a,strong
'impact on producers' cropping decisions.:

SThe destabilization of certain world commodity prices as a result
of the new U.S. low-reserve policy, is a farmers in Southeast
Asia as well as elsewhere. But the effect is more severe here since
small farmers are less equipped to bear risk. The rational response of
small farmers to price risks is to plan conservatively, say, by growing
mainly their food needs, so that they wauld not be forced into the cash
commodity market at an inopportune time. This can be expected to result
in a sub-optimum use of resources and a net loss in product.


S The main characteristics of'the economic setting in Southeast Asia
that have been presented are small farm size, dispersed production, a low
real wage rate in certain areas but a rising wage in others, an increasing-
ly large research and extension capacity, generally high interest rates
but availability of low-interest loans for high-priority projects, and a
strong demand for agricultural produce but increasingly unstable prices.
These factors are related to cropping systems.

Dispersed production

*Highly dispersed production can quite efficiently supply a similarly
dispersed pattern of consumption, such as that which results from high
population I'ensities. But such'structures are less responsive to new
crop possibilities, since local demand for the crops must develop along
with production. The alternative' of shipping crops to terminals that
are'linked to external points of consumption, is hampered by the high
costs oficollecting crops.; -

- 7 -

The problem:may not be as diffidult.:to solve as it first appears.
The.dbvetloptheati iu severalri Southeas t7Asian, countries .of an, acttye local
tradeb' in iwelotsq j!.green ornt,'yambeanY. rsybean, and other produce shows
the:.' rpid g.rbowth: in: local:: demand for new drops.: New; feedgrain crops
have beben'xsebd by; local livestock industries in :,:: .L. Thailand where
c4'rh has become a major crop-both for expott. and.domestice, feed. mills.


An increase in the labor wage, with industrialization, has no pre-
dictable impact on cropping patterns, Mhile cheap labor assisted. the,
early development of intensive agriculture, mechanization is also con-
sistent with a high cropping index, as demonstrated in East Asia. Labor-
is only one cost of production. The returns, say, from producing a, high-
valued crop, perhaps utilizing land in a period when it has a low oppor-
tunigy cost, may well be sufficient to encourage production even under
conditions of high labor-costs. Ditch and dike vegetable cultivation can
be seen in eveiithe high-wage areas of Southeast Asia, for example.
Furthermore, the high prices for field crops, such as soybeans,. that have
usually been grown by land-extensive technology, may":now support a more
intensive technology. '.'

Human capital ... .. :

Higher levels of human capital in the farm sector, particularly
when manifest in the managerial capacity of farmers, increase the
potential for intensive resource utilization without resort to reduction
of the size of holdings. The increasing number of extension workers and
researchers in Southeast Asia, .presumably will lead tb high levels of
on-farm human capital through training and advisory services.

Demand .

The possible impact of a stronger demand ,.for ,agricultural commodities
has been mentioned. i Higher prices will :permit more intensive cultivation,
but the mix of crops may also shift. Since .1973mthe prices, of ;grain--
legumes in the Philippines have increased about 80 percent while cereal -
grains cost only about 30 percent higher. A consequent shift in the com-
ponents of cropping patterns, and advisedly in the attention-;of.researchers,
may result.

Little has been said of conditions in cash-input markets, 'and.per-
haps little can be:found that 'is ,unique;about ;the'relationship;.;of cash.
inputs to cropping systems in SoutheastAsia-. The universal- case is.;that
.. .7 .".. '.i ;

- 8 .

with much higher prices of chemical inputs resulting from the. oil
,crisis and with 'expanding crop production, chemical-saving technology
::now has: higher return.' The biological control of insect and weed
pests that is possible through appropriate design of cropping patterns
is more prdfitable-"than- before. More efficient use, of fertilizer
' throughh sequential planting is also now a more important consideration
in'the design of cropping patterns.



In non-irrigated areas in Tropical Asia cropping patterns are
determined primarily by duration and amount of rainfall, with most of
the area subject to a monsoon climate. The very dry areas of India -
that have less than 1,000 mm of rain per year have sorghum and millet-
based patterns. Those that have intermediate rainfall grow corn or
wheat. Corn-based patterns best fit the zone of 1,000 to 1,500 mm annual
rainfall. The dridst portions of Southeast Asia fit this category. Rice-
based patterns require an annual rainfall of 1,500 mm or more and at least
200 nm/month for 3 consecutive months rainfall. Most of eastern Asia
falls into this latter category. Much of Northeast Thailand, having
slightly less than 1,500 mm annual rainfall and with 2 to 3 months of
200 mm/month is a marginal area for rainfed rice; a portion of the area
is not planted each year because of lack of water.

Beyond this very general summary, the type of rice culture or the
type and intensity of cropping pattern depends on the specific.'charac-
teristics of the rainfall curve. A rainfall classification scheme, based
on cropping potential (Table 5), has been developed for Asia (Coulter
a,et al., 1973). :The amount and the duration of rainfall determine the-
number iand, type of crops possible:. The gradual or rapid onset, or
decline, of heavy rains determines the type of culture for rice, which
crops should be used as second or third crops in the pattern, and the
relative difficulty oftillage operations. Bangladesh, Indonesia, and
the Philippines are being mapped using this system. The portions of'
land in Bangladesh, the Philippines, and the Island of Java that fall
under each major rainfall zone are shown in Table 6.. Details of zonanl
cropping pattern adaptation are being worked out through the Asian
cropping systems research network..

Irrigation '

Most of Asia's irrigation is "partial" irrigation, with water being
available for only part of the year. Because rivers are short and
valleys are steep-in much of the area;, large atdr storage dams are feasible
only in selected locations. irrigation only in the. wet season is valuable,
-however, in reducing-:crop risk in an' uncertain monsoon climate. Higher
levels of crop management and inputs can .be justified with some guarantee

- 9 -

ofavailable water. ..The partially irrigated areas of Southeast Asia now
offer great, potential for increasing crop intensity as other crops are...
added toaa.single rice crop .to make better use of available water in a.-:
growing season lengthened slightly by irrigation.

A summary of irrigated dreas in the region is available (Table 7),
but figures on duration.and quality of; water in those systems are not .,
available. The last column in Table 7 shows the,irrigated, land in;
Asian countries as a percentage of arable land. Southeast Asia has the
lowest percentage of irrigated land in Asia, 19 percent, compared with
l,21percent in South Asia.and 47 percent in East Asia. Fifty-four percent
of. therirrigated land in Southeast. Asia, is in Indonesia.

The fully irrigated, areas (year-round) already have a high cropping
intensity. Since multiple cropping of rice depends upon a reliable
-water supply, irrigation and multiple cropping show a high statistical
relationship. East Asia,.with the largest percentage of arable land
under irrigation, has high indexes of cropping intensity ofbetween~150
and 200 (Chao,:1975). The irrigated areas of India (Rao, 1975),.Indonesia
(Oshima, 1973)., and Thailand (Manu, 1975) have moderately high cropping
indexes of, b.twepn 125 and 150, and countrywide the Philippines has little
irrigation and a cropping intensity index nearer 100.

However if rice data are removed, the relationship would probably -
not be so clearly demonstrated...,Highly intensive cultivation.of unirri-
...gated upland crops is found in several countries The Gunung. Kedil area
of Java and Batangas Province of the Philippines are examples of such
cultivation, .. .


.Soil conditions influence cropping patterns- primarily from the stand-
point of water. movement.and drainage and tillage capability under high
-.rainfall. 1Many soils of the high-rainfall areas are high in clay fraction
and are predominantly montmorillonite in content... :The paddy rice system
,-is.the only feasible alternative for most of these areas during the p.ek
of the monsoon, when the soil cannot be tilled and maintained in an up-
pland conditior ..,.In upland areas with rolling topography, upland rice-
may be grown onl- where the soil has good internal drainage. The suita-
bility of fresh water alluvial areas (having a relatively high calcium
level) to upland crops like sugarcane hinges on this better internal
-water movement. The north shore of Java is thus suited to cane, while
:.; seemingly'similar areas in the Bangkok flood plain and Mekong-Delta, being
tmarine'ini origin, have higher sodium content, less internal-water move-;
-ment.,:iand more -limited potential for non-rice crops. The-re soil water.
movement is limited the paddy system is used.
The potential for growing a second ',r third upland crop after paddy
rice in areas where rainfall is not adequate for additional rice'hinges
on the soil's capability of being converted from-the puddled to an upland
condition.:. If the'soil is alluvial with a relatively high silt fraction,
"the puddled soil can often:" be tilled at a relatively high moisture level
Sto achiev6e structure :for upland crops, The 1:1 clay types (red soils) are
included in this..category. The more common heavy 2:1 clay types are more

!-. .. 7

- 10 -

difficult to convert. They must be tilled dry(requiring considerably
more power), then allowed to alternately wet and dry, permitting
shrinking and swelling to fracture the large clods and regain structure
for upland crops. This process cannot be accomplished in periods of
high rainfall. The rotation of upland crops with paddy rice is thus a
complex practice that requires a careful blend of power, tillage methods,
and choice of crop to fit the specific environment.- With continuous
irrigation it is thus far easier to maintain a rice-only pattern. Where
water is limiting, however, sequences of rice and upland crops make more
efficient use of the water resource. Here the tillage problem is a major
constraint to high crop productivity. IRRI cropping systems research is
concentrated on this kind of mixed pattern.

For upland crops on heavy soil, tillage for land preparation and weed
control is likewise difficult during periods of high rainfall. This-is one
of the reasons why sugarcane or tred crops are suitable on the well-drained,
heavy soils. Mid-season tillage requirements for these crops are minimal,
with land preparation being done primarily in the dry season.

Native soil fertility is a determinant of cropping patterns, primarily
in small farm agriculture where cash inputs are scarce or unavailable.
Continuous cropping, especiallywith-cereals, requires high levels of
nutrient inputs, unlike the one-crop-a-year patterns which in the high
rainfall tropics can do reasonably well without added nutrients (Geertz,
1963). Single-crop paddy rice is especially efficient in using available
nutrients under anaerobic soil conditions of the paddy. Low cash-flow
farming systems can achieve the nutrient inputs needed in crop areas by
the cycling of human and animal wastes, as in the hills of Nepal or in
parts of China. The cycling of human and animal wastes in rural areas is -
a labor-intensive process, however, and works well only under certain socio-
economic and physical conditions. The cycling of urban wastes may have
more potential in the future. The lower requirements for applied nutrients
in the cereal-legume sequences make them more desirable to the farmer with
limited access to cash inputs.
areas with
In Thailand and in/similar soils the higher fertility requirements of
paddy-upland crop rotations on low-fertility soils are limiting to crop
intensification and diversification. The favorable nutrient availability in
the paddy leaves the following upland crop with a depleted nutrient status
which may be uneconomical to correct for many upland crops. The transi-
tional soils bordering the Bangkok flood plain fall into this category and
have limited potential for cropping with upland field crops after rice,
because of both nutrient and drainage problems (Anonymous, 1975).

Elevation and temperature
For most of Southeast Asia the relatively small arable areas of higher
elevation are used forthe production of vegetable crops. At higher latitudes
and on the drier Deccar plateau of India, winter temperatures influence the
crop pattern, but year-round cropping is not prohibited by temperature in
most areas.

'The need for agro-climatic classification
A great deal of empirical information on cropping pattern potential
is available for Asia. Coordinated efforts are now underway to utilize
existing information on soils, 'climate, socio-economic conditions, and crop
adaptability to map areas of equal cropping pattern potential in Asia. The
climatic map of Asia now in progress is one link in this effort (Coulter et
al., 1974). Preliminary estimates are available for Thailand (Sriplung, 1973),
portions of India and the Philippines (Philipson et al., 1972), Sri Lanka
(Panabokhe and Walagama, 1974) and for Southeast Asia as a whole (Obradovich,

- 11 -


The homestead area

Asian farms are characterized by a diversity of crop, animal,
and off-farm enterprises which contribute to the'cash flow of the
farming system, and a homestead iprduct ion area whichh is aimed'
primarily at farm family comfort and subsistence. Major crop
enterprises will be dealt with individually. Attention will be given
here to the homestead area.

The.,area around the house or farmyard is normally pla ted to a
wide assortment of crops which not only offer :'sh'iter and privacy, b'ut
also contribute diversity and quality to the die'of the farm
Therelative importance of this the family"'depends on both
farm size and its cash flow. If the family has a low income, the
extent and quality of the homestead area i's crucialto the.quality.
of the diet. Once the family's requiremnnt for staple";food h'as been
met, the development of the homestead area is the next step in improving
farmer welfare. For a subsistence farmer the homestead is 'as important
as his cultivated field area. The development of homestead areas is'
most advanced on the small, low-income farms of Indonesia and Nepal.'
These farms are characterized by a diversity of economic plant species,
which number 50 to 60 in the more .advanced systems. The plant
components of the system may include: 5 or 6 tall-growing tree species
(coconut or fruit), 5 or 6 1me.dium-4 ight tree species, 5 or 6 bush or
shrub species, 4 or 5 root crops, and up to 30 shade-t'olerani, short-
statured or vine-type annuals.

These crops provide a wide diversity 't. the diet. Thy require.
no purchased inputs. Most of them require low management. Their.
growth .is.usually luxuriant because of the inflow and accumulation of
nutrients, n the homestead area from human and animal wastes.:; This
use of,.accumulated nutrients in the homestead area is in fact an
efficient recycling system which does much to offset the frequent
lack of conscious recycling into the crop area. Hedgerows surrounding
field plots can be considered an extension of the homestead area.
Both areas, having a high portion of perennial crops, have relatively
stable productivity.

These factors limit the development of homestead areas on low
income farms: i presence of untended forage animals which severely
. limit crop diversity (especially in India),: 2) clustered rural
villages, 3) seasonally deep waterr, 4) a shifting village site,
5) high altitude, and 6) location of the village in an estate crop
area'where space for alternate crops is not made available.

- 'I ..~' I

While little has been recorded concerning the content and importance
of homestead areas, the effects of their presence or absence can be
readily observed in rural households. Low-income people who have to
purchase all their dietary needs or who produce their own staple only
will have a much lower dietary standard than those having access to a
highly diversified and stable cropping system. This key component of,
the farming system has until how been overlooked in efforts to improve
the lot of the low-income farmer. The high management-requiring vege-
table crops of the often-cited "kitchen garden"' efforts are an ineffect-
ive substitute for a well-designed homestead area.

Large farms (above 5 ha)

Because of their unique (for Asia) makeup, larger farms are
briefly mentioned here as a separate category. Little will thus be
said about their.operation. They are characterized in general by a
more "commercialized" system of management that has usually a source of
mechanical power, market orientation (a relatively small portion of the
produce being consumed on the farm), relatively high cash inputs, hired
labor, and higher capital investment.

The cropping patterns are estate crops or simple crop.sequences
of a relatively few crops. In terms of number of farms, this type is
of. minor significance in most countries, but. in terms of national
economy and production they are much more prominent. Most of the so-
called "modern" (capital, cash, and power intensive) technology is
immediately relevant to this kind of farm. In-most Asian countries,
however, population pressure on land precludes the consolidation of
smaller units into this .type. of system.

The crops grown on such farms are: estate crops (sugarcane, soil
palm, rubber, coconut, tea, etc.), timber, pineapple, corn (Southeast
Asia), wheat (India, Pakistan), and rice (to a.very limited extent).
Crop patterns/are saimple-crop sequences, of relatively few crops
that have a relaxed planting schedule between crops. The future ..
development of these large farms -- which revolves around varietal
improvement and the creation of a more efficient infrastructure to
service them -- icj:not dealt with here.

Animal-based systems

Animal and crop production interact at several points. Crops
and crop residue provide feed, and the animals provide power. Animal
and crop enterprises usually compete,, but they are complement-
ary in the case where livestock are raised within crops, say coconut
groves. Animals have uses which further complicate the analysis.
They are used for ceat, milk, as stores of value,, for recreation,
and are a source of status and pride. Bovine husbandry for the purpose
of power would, in many cases, not be profitable were it not for the
non-crop-related livestock uses.

- 12 -

- 13 -

,The.,use of cattle for plowing in the East Java region around
,Probolinggo is an instructive example of some of the relationships
.involved. A surplus.of.cattle is raised here and shipped to other
areas for slaughter. Many landless people work as caretakers for .the
small household herds. The animals are used for field work, even
though the wage rate is one of the lowest in Asia -- 0 to 20 cents per
, by one account.

The area is quite dry -- receiving about 1,.0,00 mm of rainfall a
:year, most of it during 4 months. In this aspect the region is similar
to Northeast.Thailand, also a major bovine exporting region, that-has a
long dry period during which crops cannot be grown. Similarity of the
regions suggests that if a resource base includes a long dry period,
an optimum cropping system will likely include bovine production.

Another aspect of Probolinggo cattle production is that the
animals are highly prized and the bulls are used for racing. This
explains why the livestock enterprise has flourished even though
the average farm size is well below 0.5 hectares, unlike in Northeast
Thailand where farm sizes-average above 4 ha and soils are heavy,
In Probolinggo tht medium-textured soil could easily be tilled by hoe, as
is done elsewhere in Java.

Finally, another aspect of livestock production in East Java is
that the Maduranese migrants brought with them their breeds and a
highly developed technology for cattle-raising. By comparison, the
Javanese migrants who have settled in Lampung, Southern Sumatra,
brought with them no such capital. Bovine production and utilization
in Lampung is still a weak enterprise despite larger farm sizes,
fallow tracts, and deficiency of labor. A few cattle are raised, but
not used for plowing, most probably because bovine technology is still

The simple interaction of bovines and crop cultivation, that is,
where bovines are fed on miscellaneous vegetation and crop residues
and used mainly for power and later salvaged for meat, is.a frequent
component of cropping systems in Southeast Asia. But in different
areas even this simple type of interaction needs fine adjustment to
special conditions that often are related to feed supply.

In areas where the landscape is mostly covered by lowland rice
paddies, rice straw must be collected and stored for use when crops
are growing. This may determine the way in which grain is harvested
and threshed. Mechanical threshing which impairs handling or
storability of straw is not suitable here. In the Thai Central Plain,
most rice straw is harvested and can be purchased and sold, but straw
has no value in other parts of the country where upland forage is

14 -

F ..armers', in Batangas,.Philippines, intercrop a few rows of corn
with. upland .rice to provide';feed 'to draft animals during the growing
season-; This is perhaps necessary .because crops cover the-landscape
..:almost throughout: the:year and the oone :'rop of rice straw is not a
year's feed supply.
-. ,, *-- ?'
Tractors and draft animals are closely competitive in many areas
and it is the minor interactions that may push systems toward one power
source or.the other. Restriction of feed supply, say, by intensity
.,; ;in crops which donot produce palatable residues, is a factor of some
;importance-to the viability of animal-based systems. :Increased demand
..:for meat as.incomes rise, on the other hand, may encourage their
.expansion :.. : . .-

Shifting cultivation ;

Aside from the larger: "capital intensive" farms,: most :other
Systems. can best be viewed in terms of a dynamic process of change,
with different stages of development in the same system seen through
examples from different countries. 'There are three basic environment-
dependent development patterns in Asian shifting cultivation systems.
All are limited by a lack of natural production resources; some, by
a..lack of available market.and :input resources. All arise" from the
population pressure on"'the better agricultural land that has forced
landless .settlers: onto land with marginal natural production capability.

These areas are: 1) low-rainfall areas of flat to gently rolling
topography with soils of very low natural fertility, typified by the
"Chaina" areas of' Sri Lanka; *2) low-rainfall mountainous areas,
typical:of western Nepal; and 3) high-rainfall mountainous areas.

The population pressure that forced movement into the less
suited agricultural areas often precludes planned movement out of them.
Land often is simply not available. Shifting cultivation in Asia has
J.een.described in several major works (Barran, 1958; Conklin, 1954;
Davs,; 1973;. Moerman, 1969; Spencer, 1966; Reed, 1965).

NearLy all.workers remark on the ingenuity of peasant farmers
and their propensity toward experimentation on at least a small
portion of their farm, accounting for the evolution of efficient
systems .Davis (1973). states: "Ifraefficiency is measured in terms
of the: ratio between labor: inputs-and yields, rather thari'in terms
of gross production per-unitc landd: swiddeh (shifting cu-ttivation)
often is- airemarkably 'efficient 'cultivation system,' ingeniously
.,.adapted' to, low-energy :techrrologie's' in -forested regions". We cannot,
however, be s;atisf ied: withJ Il6, productivity per unit of land area;
and population pressure is rapidly forcing the slash-and-burn' '.'
cultivator into an ever less stable system.

Development and stabilization of shifting areas takes several
pathways. All include intensive development of the homestead areas.

- 15 -

The relatively dry, low-fertility, gently rolling areas of Sri
Lanka are stabilized. as.follows: .

Starting point: Shifting cultivation with 2 or 3 years of cropping-
and a 10-year fallow no farm power.

Development phase 2

a. Construction of communal water catchment tanks with production
of the staple crop below the tank. Low soil fertility makes
paddy rice more productive in spite of limited water.

b. Addition of animal power.

c. Continued shifting cultivation above the tank, Farmers have
cultivation rights in both areas.

d. Development of the homestead area with tree plantings.

Phase 3 ....

a. Markets and cash inputs available -

,b. Cattle raising added as an enterprise '

.:. c. Fertilizer inputs below the tank, making upland crop production
possible using less water. '-

d. Pasture, low intensity annual-tree interplant cultivation
above the tank to stabilize the shifting portion.

e. Gradual entrance into a cash economy.

This process progresses by itself slowly, but research efforts
now underway should hasten it. Water recycling studies at the Inter-
national Center for Research in Semi-Arid Tropics (ICRISAT) in India
should add considerably to the knowledge of how to build water catch-
ment systems. Technology: 6r ow intensity cropping'on low fertility
soils under conditions 6f loiwrainfall combined with animall production
systems has been perfected in the dry portions of South Central Java.
Here a range of dry zone tree crops, some serving primarily as a fuel
source, are combined with such drought-tolerant crops as sorghum,
cassava, and crotalaria in intercrop and relay.combinations.

We suggest, than, that to stabilize this kind of shifting system,
proven traditional methods from similar.areas in a more advanced,;stage
of development, be combined with new varieties and modern inputs to give
more efficient ue odf'available sourcese, higher productivity, and a
more, varied and stable source-of' food for,.the ;:farm family. .We suggest
for this, as for other systems, "The use' of new.inputs in combination
with'traditional production factors in a new technical relationship.--
a movement to a new production response surface" (Hopper, 1975).

.-. : -

Shifting systems in the high-rainfall mountainous.areas can be
stabilized following the pattern in the Himalayan foothills of eastern

Starting point: Slash and burn with no farm power. Two or three years
cultivation in a 7 or 8-year cycle.

Development phase. I:- .Establishment of terraced rice paddies in the
:.. lower, regions. Addition, of.orage-draft animals.

Phase 2: Limited cropping of non-terraced upland areas. Extensive
development of the homestead area.

Phase,. 3;:Labor-intensive cycling of nutrientsis begun if the system
has no access to markets....(Fig.. 2).. This eastern Nepal system
represents the most advanced stage of purely traditional
technology yet found inAsia., With market availability,
purchased fertilizer inputs partially substitute for the nutrient
cycling. The extremely intensive but completely subsistence-
oriented systems of the eastern Nepal foothills represent the''
ultimate in this development without market availability. With
Markets available, a stable and highly productive (but labor-
intensive) system is possible. This system.would afford the
farmer who has no more than 2 ha of land a high-quality diet,
stable production,andmoadest .(by Asian standards) cash income that
r'ould enable him to obtain the fewnecessities for existence.
It would not provide large marketable surpluses of food nor
anything 'approaching.a western standard of. living, however.

Shifting cultivation in the drier mountainous areas, such as
western Nepal, can move, only..toward an.extensive tree-oriented homestead
area and controlled cattle grazing with some pasture improvement.

S :::None of these systems,cqen afford the capital and cash.input
requirementepofmechanical power.,..The forage capacity of draught
and.slaugbter animals not only..adds considerably ;to the productive
efficiency of these systems but also minimizes cash output for power,
.The ,limited. cash resources, can be used for fertilizer inputsl to correct
the primary., limitation tpohigher., productivity in a shifting system.

.-pma.steLd crop tephnglogy is..critical, as is a knowledge of tree
and annual-c.rop relay and intercrop. techniques.

;Rolling.,upland areas ... ., :

In Southeast Asia-the low-rainfall upland areas that have less
than 3 months, of- rainfall.:above 200 mm/month are devoted primarily to
corn production ,::9n frnso around 2 ha draght..animals are used'..
A .we developed, upland, corn farm yill supplement animal with tractor
power. for'.priiary 1iage It W, ;have an extensive homestead area,

- 17 -

and fence rows planted on .the contour for control of seasonally heavy
rains. It will combine animal-fattening with its draft.
animal source and will use fertilizer inputs.

With higher rainfall (5 to 6 months with 200 mm/month or: more)
the system will be based on upland rice.. -.;

A time profile of such system seen in the following: ,

Initial stage: Settling and initial land clearing with no animal
component or power source. Intensive intercropping of annuals
having different growth stages is used based on cassava, rice,
and corn mixtures. This gives an 11-month growing period
requiring a single primary tillage operation. Nofertilizer,,;
-inputs are used. This is typical of the transmigration area:..;:
of South Sumatra in Indonesia. .,.

Stage 2: Draft and slaughter animals are introduced which utilize
available feed and provide .power. The cropping pattern,changes
to the use of.several different crop sequences. This spreads the
power requirement but uses power mostly for primary tillage.
This is the shift toward a power-based system,. The homeatea.:and
fencerow areas are then developed.

Stage 3. Contract tractor operations .begin to handle some of the
primary tillage, with animal power providing seedbed preparation
and weed control as well as primary tillage of small-plots and
wet fields during high rainfall. Intensive use of chemical
''fertilizer becomes coicmon. Cropping patterns, on a given farm
will be varied from.field to optimizee labor and market
resources (Grimble, 1973). Field..and then garden vegetable; ..
culture is introduced as labor availability increases
appear. Intercropping corn with upland rice or other.crops-is
practiced in intensive cropping. The IRRI outreach research
location in Batangas.Philippines is typical of this latterstage
gsi reported by De Ocampo et al. (1969). IRRI seminar reports
'have recently described this system in detail (Herrera and Harwood,
1.975; Samson et al., 1975, Garrity et al., 1975). A future
publication will fully,,report these findings. ,A similar site is.
being established as part of the Asian cropping systems network-
in Sumatra by the Indonesian national program to study phases 1
and 2 of this system and to attempt to hasten the occurrence. of-.
phase 3.

T'he well-developed homestead area, the mixtures of upland crops,
the intercrdppidg of trees (coconuts) with annuals, and the integration
of animal enterprises into the system are critical aspects of this
system. Ready markets for field crops and an available source of
fertilizer are absolutely essential. Improved crop varieties of suitable
plant type and precise maturity requirements, along with rather complex
pest management practices are also necessary. This has all been put

- 18 -

together by the Philipino farmer in Batangas Province into one of the
most effective combinations of traditional patterns.and modern technology
yet found in Asia.

This system follows the same patterns as those found in the drier
upland areas, but uses shorter duration crops in intensive sequences
and has far greater diversity of crop and animal enterprises. Similar
intensive integrated animal-crop systems exist in Taiwan (Lee, et al.,
1973) and in Malaysia (Ho, 1962).

Fully irrigated systems

With adequate water for year-round irrigation the farming system
in East and Southeast Asia on small farms will be based either on con-
tinuous rice (5 crops in 2"years) or on continuous high-value crops.
Continuous rice is most commonly grdwn for several reasons. Most Asian
irrigation systems have been constructed in the low-lying areas .that
tend to have heavier soils and a high water table, simply because
:" .. ater is easy to get intd the system, less lost and
construction costs are cheaper. The systems are designed. for irrigating
paddy rice, with low-cost paddy-to-paddy field distribution, and
surface drainage only. Below-ground-level drainage is not needed for
paddy rice. It is therefore much easier to continuously puddle the
soil forsequential plantings of lowland rice. Land preparation inthe
paddy requires low power inputs, and irrigation to achieve standing water
in'the paddy requires far less precision than do. the more exacting
irrigation demands of an upland crop on.clay soil. It may be true
in the dry season less water would be needed for an upland crop, but
the conversion of puddled clay soil to an upland .condition, the
construction of ridges and furrows for in-paddy drainage and the higher
fertilizer inputs required for.the upland crop all make continuous low-
land rice more attractive unless water availability becomes the primary
limiting factor.

Sugarcane is irrigated in. Indonesia and in other countries,
mainly bn large farms. In the very'small farms (less than 1 ha) that
'have access to markets, continuous vegetable production will be
followed as a third use of continuous irrigation facilities. This will
often be through the use of raised beds. The ditch-and-dike system
of Thailand is typical. It is designed around,the water control needs
of vegetables in low-lying, high-water table,areas in the alluvial
plains around the large coastal cities of Asia. This system has been
thoroughly described by Sritunya (1975). The sorjan system of Indonesia
is a modification of this method designed .or areas with a seasonally
high water table. This method uses a broad ditch for growing rice
between the raised beds on which upland crops are grown.

.19 -

Rsainfed rind: iprtia~ly irrigatedd: rice-basdd systemss

.: -:.i:-:.The setting: for; rained and partially irrigated lowland rice has
n'.,Lbeen- .described:. Theisystem accounts for the bulk of Asia's rice and
the : majority of :small farms in East and Southeast Asia. Annual
.. , generally above 1,500 mm per year, the soils are heavy,
farm.a -.-5 h-a, and the farmers sell about half of what they
produce.,. Irrigation is supplemental during the rainy season, sometimes
.,.extending. or 2 months beyond the end of the rains'. Much of this
;category-~of rice-based system is found in rainfall zones whdre th'
monqthly rainfall is 200 mm per month for 5 to 6 months a year. In most
:,.instances-,. a single crop only is harvested, even with partial irrigation.
Because these water-limiting situations offer great potential for more
.ntepsive .cropping, they:are the major focus of 'IRRI crpping systems
,r..2esearch. .; In most years.'two crops of .rice,'or attA leat ride followed
i. by an uplan cprop :are possible. Our e&search:is'largely-aimed at this
S.largq sector of- Asian,agriculture. .* Severallapproaches are Bingg used.
,Early,-maturing rice varieties are :being directt seeded-in non-puddled
soil ;ati:the start ,of -the hains iniareas whire th6 mbnsoon s afts gradually
;.- after.a dry period ..:This method works only where thesie~aesi i be''::
prepared at least amdonth before tthei rains begin, and where weed':
:populations are:.not -too:dense.. If' thd rains :egin suddenly rather 'than
gradually :(climatic -s'ubiones.-3.2 rather than .3il) the soil ms st be''puddled
a'. nd.-pregerminated seed sown onrthe:s-o ft urface' rather"than dry sowing.
Ti-;:,.This because .rice-will- :not :-emge rf.romnbeneath the soil. under conditions
of high oxygen tension, ".-These4 directt seeding methods replace the tradi-
tional',long wait until- the monsoon is well established, then puddling.and
transplanting. This l:at-ter.method uses the 5 months of good rainfall for
a::ssingle rice crop. lDirec't seeding permits :the growing 'of two short-season
.. rice .crops during, the rainy period, "iwth 'ih second being transplanted in
, the normal fashion.

Upland crops after rice depend on soil texture, water status and
tillage capability (Harwood, 1975a). Change in soil bulk density with
puddling and tillage capability unde'r upland conditions are closely
related to multiple-cropping potential (Fig. 3). Low-tillage banage-
ment following paddy rice offers a wide range of options for several
.crpps .

Partially irrigated areas normally have a somewhat more dependable
- water source during. the miinsoon but nevertheless still run out.. water
shortly after the&.rains.end. Two crops o:f: riie dfl'lowed by a short-
season drought-tol'erant crop like ighreti ui ta) are
often possible. .


Ctf" sequences .. .

Sequences of crops grown in pure stands.-,are-bhy':far the ost :st :
common form of multiple cropping. ,,In!,areas wjt.h a-coolfwinfery"'paddy :
rice-wheat (in the high rainfall areas) is grown. This would include

20 -

such areas as the low-hills of Nepal, northern India, Pakistan, north-
western Bangladesh, Korea, and portions of the Peoples Republic of China.
Most of these areas have soils that are courser in texture and easier to
till than those of Southeast Asia, making the paddy rice-wheat conversion
much easier. Maize-wheat rotations are used where summer rainfall is -
moderately intensive. The drier' regions of India have multiple crop se-
q-.ences as reported in several comprehensivee summaries (Anonymous, 1967,
1970; Nair.'et al., 1973). Similar summaries are available for Sri Lanka
(Upasena and Fernando, :1973), Bangladesh, (Carolus and Kaza, 1958), and
Taiwan (Iso, 1'954; Cheng,'1972). Little attempt has been made to relate
these crop sequences to physical or socio-economic conditions of their
:respective areas, so generalizations are difficult from existing reports.

The suitability' of specific crops to the more common conditions is
generally recognized, however. The soil requirements of most field crops
are well known to all of us. It might be worth mentioning, however, that
for the puddled, heavy clay soils following rice, where tillage is diffi-
cult or impossible, the soybean is probably the most well-adapted crop.
-F' r these same conditions the vine crops such as watermelon, cucumber
squash,. ampalaya, or the viny type of legume are well suited, since plants
are spaced at wide intervals and a suitable seedbed can be prepared by
digging holes at spaced intervals for hill-planting. With zero tillage
after paddy rice these vine crops cover the soil before weeds emerge.
The moisture requirements of various crops also determine their places in
the pattern. In a typical:monsoon rainfall pattern there may be a 4-month
dry period with less than 50 mm of monthly rain, then 1 or 2 months of
100 to 200 mm/month, 5 months of 200 to 300 mm/month,"-one month of 100 to
200, then the dry season again. In auch a pattern dry-seeded upland rice
fits only at the start of the rains, since it must germinate before the
soil becomes saturated by heavy rains. Rice can be harvested wet, so heavy
rains at harvest time are less damaging to rice than to most other crops.
Corn can likewise be planted early, as it can be harvested in wet weather,
especially if it is for green "cooking" corn. The legumes and sorghum
cannot be harvested in wet weather,-and fit better as a second crop at' the
end of the.rains when their drought-tolerance is also'-at maximum. Crop
selection is thustbased on"both crop growth and management requirements.

For paddy rice growing areas now producing a single .:crop of'rice a
year with 5 to 6 months of high rainfall, direct-seeding followed by a
second.crop of transplanted rice can then be followed by low-tillage
broadcast mung bean or cowpea in a larger portion of the rainfed rice area.
Potentials for increase in cropping intensity in rainfed and pattially
irrigated areas by growing an additional crop in sequence are thus highly
promising. Where that crop is either paddy rice or legume, additional
fertilizer requirements are well within reason (Nair and Singh, 1971).

In upland rice areas, a single crop is now normally grown, but most
areas have the potential for at least a second crop. Here.n wide range
of field or vegetable crops is feasible. Unlike for second-crop paddy,
however; high levels of fertilizer must be used with intensive upland
crop sequences. 'If productivity is to be maintained the only alternative
is the Nepal system of labor-intensive nutrient cycling, ..

- 21 -

Tree crops intercropped with annuals

Coconut, oil'pim, and young rubber trees are often inte.t'ropped, in
the early years of plantation, to a range of annual crops. Cassava, soy-
beans, peanuts, corn, and other crops may be used. Most.research work
to date on this system has been coordinated' by the Rubbr-kesearch Insti-
tute in Kuala Lumpur, as summarized by Blencowe (1969). Smallholder tree
intercropping in the-Philippines is done-mostly'in young--maing or coconut
plantations, where upland rice and corn are frequently used.. In older
plantings the more shade-tolerant cassava, ginger, aindt er! minot'root
crops are used.

Intensive tree intercropping

Complex mixtures of trees of different canopy heights and shapes,
characteristic of homestead areas, are also.found in more extensive
plantings, especially at altitudes of 300 to 500Y- ; bre coffee grows vll
Such mixtures may include two ur three fruits, (Gliriciaia sepium) tor
firewood, and a range of lower-growing tree and shrubs.

Similar patterns are found in dry areas of Southeast Asia using
kapok and other drought-tolerant species. These plantings are charac-
terized by stability of both production and management. They require
low power and chemical input and a modest but continuod'slabor input.
They are thus ideally suited to the resource pattern of the small farmer.
Because of the.long establishment time,-however, these crop patterns are
usually heavily intercropped with short-season annual cash crops in the
early years.

Intensive vegetable production ..

The highland vegetable production, the ditch and dike, and the sdrdjan
system all utilize intensive vegetable relay and intercr6pping techniques,
separating them from most other types of patterns. These are situated near to
markets and are high cash-flow systems. High levels of insecticides, fer-
tilizers, and compost are used in extremely labor-intensive systems
(Sritunya. 1975), The intren-ive vegetable trellis system in the Philippines
is similar in resource requirements. Here permanent horizontal wire trellis
is supported 1.8 meters above the ground on wood or concrete posts such that
draft animals can be used for cultivation underneath. A year-round sequence
of climbing legumes, squash,:pumpkins, and gourds is grown.

Complex intercropping.of annual field..crops

These patterns are highly developed in Asia as reported in IRRI's
previous reports (IRRI, 1974, 1975). We will not go into detail as to the
numerous combinations and timing of crops, but will merely summarize the
types of patterns and list-possible reasons for their; use.

22 -

A. Crops of similar plant type but of different maturity
(Nigeria Andrews, 1972)

S Millet, corn.(3 months)

Sorghum. (6 months)

'B. Tall crops with shorter crops below them to establish different
:levels of: leaf canopy.

1. Short season crops under permanent (tree)crops.

Time e

Coconut, rubber, banana.-

/ Annual crops /

rainy season

2. Short-season crops planted at the start of the growing
period of long-duration crops.

sugarcane (more than 1 year)


', "< :r- ',

S.tort-season crops
(corn, soybeans)

Annual crops (less than 1 year duration)

a. Major (tall) crop harvested first

Corn (3 months)

Cassava (10 months)

Corn (2V-3 months)

Peanut, sweet potato, rice

(4 months)

Minor (short)crop harvested=

Mung(2 months) /..,.

C6rn(3-4 months)
1 I :: :: '.....;' -,'.:



- 23 -

Planting arrangement. Two types are used, depending on the permanence
of the system and on the degree of organization.

1. Mixed planting a random arrangement of each species in a given

2. Alternate-rot planting (ihterplanting) the oonfinemeit 'bf
each species bo a uniform row.

The possible reasons for intercropping are as follows:

Labor use efficiency is high when power for primary tillage is not
available (Norman, 1968). With no poer,i long-duration intercrop se-
quences-(usually based on cdssava) are used, producing three or four crops
in a 10-month period-and requiring only one primary tillage operation per
year. The rice-corn-cassava pattern 6f Indonesia is an excellent example.

IRRI data (IRRI, 1974) show thatii specific cases both insect and
weed management may be easier in intecr6p combinations (Litsinger and
Moody, 1975).

The" otal productivity of the patterns seems to be significantly
higher only lf there is-a major difference in maturity among the component
crops. RIe-.'-orn, corn-peanut, and mung-corn are examples (I1NI, 1974).
The commonly used comr-soybean pattern is not as productive as tihe-sole--
Sprop plantings. It seems probable that in an area where corn is the-iomi-
nant crop and downy mildew the major limiting factor in its yield, the wide
,(2-m) row spacing of corn in the intercrop combination may be a major factor
in reducing mildew incidence.

The non-legume combinations such as rice and corn have repeatedly
shownhigher efficiency of uptake of applied nitrogen than has either
crop alone (IRRI, 1975).

The question, of "harvest insurance" from intercropping arises in
almost every report on the subject. A careful review of the'"iterature
however, reveals no evidence to support this hypothesis whi{4i as been
proposed for map yy 'rs. Our .experience at IRRI shows thaf 'ffa crop in
a mixture is to. be.differentially eliminated or reduced in '"ffeld by some
factor, it will normally happen after considerable vegetative 'growth has
occurred (for instance rorn downy mildew infestation in the' corn'-and-rice
combination). The non-yielding crop will exert its competitive effect
without producing anything, resulting in a lower productivity'6f the
mixture than from the separate monoculture plantings. 'iat..fiaend to
show that with a single crop failure, individual monoultur 1patitings
are better. .,The desire for stability of production a' id'i:O"'b'insurance"
is a strong argument for crop diversity but not necessarily for crop
:mixtures. The.burden of proof lies with the advocates of the "insurance"
hypothesis. On close inspection, traditional patterns and systems have
arisen for positive,,,not for negative reasons.

24 -


There is no question that continued institutional and infrastructure
development to enlarge the resource base of the Asian farmer is critical.
At the same time,however, technology is needed to. make more efficient use
of present and planned farmer resources. Countless small farmers in Asia,
in the foreseeable future, will not have access to irrigation, to nearby
markets, to mechanical power, or even to easily obtainable cash'"inputs.
Our efforts at providing technology must, of course, be largely directed
toward that highly productive and more easily reached segment that .accounts
for the bulk of Asia's food production.. Social pressures, however,':are'
forcing a mandate to help the heretofore "bypassed" sector, those "small"
farmers with more limited resources. The clamor for "peasant-biased"
technology (Griffin, 1972) has thus far resulted in much noise but little
action. Norman (1970) suggests that increased productivity is possible
within traditional systems. Earlier in our discussion the pathway of
development was indicated for several of the more "backward" types. of
farming systems in Asia. The question now seems to arise: 'Can change'
(development) take place for these bypassed farmers without the complete
realization of Mosher's "progressive rural structure" (Mosher, 1969).
Present indications seem to suggest that, within limits, it can. It must
be kept in mind, however, th-at we are not talking' of transforming every-
peasant'farmer into a market-oriented, capital and cash-intensive entre-
preneur who devotes his afternoons to balancing his account books. Perhaps,
now that our "modern" technology has failed to accomplish this ambitious
goal, we should first set out to move the peasant farmer a step away from
malnutrition, disease, and starvation, then-worry about providing a surplus
cash income..

Evidence is beginning to accumulate suggesting that elements and
patterns of traditional systems may be quite efficient. Some suggest
that in nations where, because of high population pressure, production
resources are limiting, high efficiency is critical (Conway and Romm,
1972). This has been illustrated graphically (Fig. 4), showing the
higher efficiency of resource use with contributions of traditional pat-
terns with modern varieties and inputs 'typical of the N response data for
corn-rice intercropping (IRRI, 1974). This also illustrates Hopper's
"new response surface a combination of new and old."

A major thrust in reaching small farmers remains in the hands of
the plant breeders. It. is essential,-however, that at least some varie-
ties be developed tb fit.the low cash-input situations of our target
group of. low-resource farmers. .Disease and nrisect resistance may not be
as important in.trditional settings as are drought tolerance,perhaps
Shade tolerance, Vigorous plant type, and proper maturity (close analysis
will show that most traditional rice 'and corn varieties in Southeast Asia
do not, in fact, have much resistance. to -any of the major pests and
diseases). This does not argue that resistance is not important.

- 25 -

We are suggesting that the lack of pest resistance, however, is not the
main reason for their non-acceptance by many traditional farmers. Plant
breeders have a big challenge and can contribute significantly by developing
improved varieties'to fit into:the low-resource environment.

Secondly, if plant materials could be made available that would
expand aid improve the diversity of-homestead areas, major improvements
could be made&-i the diets of rural-landed peoples. In 1he high-rainfall
areas of Southeast Asia there is little real excuse for the persistence-
of vitamin-A deficiency. It will not be corrected, however, by importa-
tion of carrot seeds. High-management crops are not readily accepted'"
into low-management 'systems.

This argument points to several needs and approaches which only
recently have come into use. Heady and Agrawal (l~97) suggested the
farming systems analysis approach to multiple.cropping. Spedding (1971)
furthers the notion that the approach should be based on an ecological
awareness and a framework that makeI ptimum use of available' resources.
It should be realized, however, that, and capital-intensive
technology has failed, systems models based on these same principles can
hardly be expected to do much better.

The approach being used at IRRI is to come to a' understanding of the
existing system before trying to determine how to change it. This is es-
pe'iaty 1dritical with traditional systems that are completely alien to
our "western" mentality and education. The systems,~approach is critical,
since in traditional systems the components are closely integrated. The
study, to be successful, however, must,start from the "ground up." It
must end in an understanding of the system and have a relevant technology
to offer.

A further example of this "relevant" technology are the low-cost
power tillers designed to better fit the resource pattern and needs of
the small farmer with an intensive cropping systems. Small multiptipose
threshers and small planters are other examples. Such power units may
be critical to the efficient use of production resources in intensive
cropping systems (Bradfield, 1970; Banta, 1973).

An Asian cropping systems network is being established to provide
-a series of research sites .(Fig. 5) to investigate possible increases in
crop intensity in likely target areas (Carangal,;1975). On these sites
a farmer-oriented research approach is being implemented to achieve data
on relevant and potential cropping systems (Harwood, 1975a). A "think
.tanklc5of research program heads in a recent mniting at Ii't 'proposed a
research model to be;followed in this research (Figure'6). --It was
recognized that comprehensive studies 'of al' thei suggested 'paimeters
are not feasible, but that, at the rminiimum an intuitive knoleddge of
interacting factors is essential. It was stated that a single researcher,
working alone,) can follow this method. The farmer involvement it:i 're'-
search is based on Johnson's (1972) hypothesis that "experimentation
is as natural as conformity in traditional communities". To date our
experience supports, this hypothesis.
Y ., .,.

26 -


Amed, K. 1965. Agriculture in East Pakistan. M/S Amed Brothers
Publications, Dacca.

Andrews, D.J. 1972. Intercropping with guineacorn a biological
cooperative: Part I. Samaru Agric. Newsl. 14(2):20-22.

Anonymous, 1967. Symposium on new cropping patterns. New Delhi.

Anonymous, 1970. Report of the National Seminar on Multiple Cropping.
New Delhi.,260 p.

Anonymous, 1975. Preliminary report of"the 1973-1974 Chao Phya
research project. Ministry of Agriculture andsCooperatives,
Bangkok /unpublished/.

.Banta, G. R. 1973. Mechanization, labor, and time in multiple cropping.
Agric. Mechan. Asia IV(1):27-30.

Barran, J. 1958. Subsistence agriculture in Melanesia: Honolulu.
B.P. Bishop Mus. Bull. No. 219.

Blencowe, J.W. 1969. Crop diversification in Malaysia: Kuala Lumpur.
Conf. Nov. 10-12, 1969.

Boyce, J.K., and R.E. Evenson. 1975. Agricultural Research and
Extension Systems. Dept. of Agric. Economics, Univ. of the
Philippines at Los Banos and The Agricultural Development
Council Inc.

Bradfield, R., 1970. Concept of multiple cropping on paddy soils
.,Paper presented at FAO/Ceylon Nationa1 Agricultural Extension
Workshop, In-Service Training Institute, Gannoruwa, Peradeniya

Carangal, V.R. 1975. The International Rice Research Institute
SCropping Systems network. Asian Cropping systems workshop.
The International Rice Research Institute, Los Banos, Philippines
: -. L/in press/

Chao, C. 1975. Improvements for increasing cropping intensity of
paddy fields in the past five years. Cropping Systems
Workshop, The..International Rice Research Institute, Los Bafos,
Philippines, March 18-20, 1975.

Carolus,.R.L. and Q.H. Kazi. 1968. Observations and trials with
..various crops during the Boro Comilla (1967-1968).
Comilla, Pakistan Academy-for .Rural Development.

Cheng, C. 1972. Multiple cropping in Taiwan. Ext. Bull. No. 15
ASPAC Food & Fert. Technology Center.

- 27 -

Conklin, H. ;'" 1954.' 'JAn ethnoecological approach to shifting
agriculture. Transactions of the New York Academy of Science

eon way,;G.; and'J;iRomi'. 1972. Ecology and resou.te development in
Southeast Asias The'Ford Foundation, Office for Southeast Asia,

Coulter, J.K., J. F. Derting, L.R. Oldeman, M. M. Obradovich, and
T.B. Slattery. 1974. An agro-climatic classification for
J'7evaluating cropping systems pot6ntitls"in Southeast .Aian rice
growing regions. .Int. Rice Res. Inst. Los Bafos, Philippines.

Davis, WiGtc. 1973. Social relations in'a 'Philippine market. Univ. of
California Press.

.iGarrity, D.G., B. O. Buenaseda, R.R. Harood and E. C. Price. 1975.
Evaluating alternative field crop"patterns -for upland rice areas.
IRRI Saturday Seminar paper, Los Bafos, Philippines.

Geertz, C. 1963. Agricultural involution. The process pf ecological
; change in Indinesia. Univ. of -'California Press. Berkeley and
.': Los Angeles.

Griffin, K. 1972. Economic aspects of technological change in the
rural area df:monsoon Asia. 69. 'p.

Grimble, R.J. 1973. The central highlands of Thailand: a study of
farming systems. Overseas Dev. Adm. specialreport. Wye College
University'f- London.

Harwood, R.R. 1974."Farming systems' n' hill agriculture. In: Moseman
A.H. (ed.) A study of hill agriculture in Nepal.: The Rockefeller

Harwood, R.R. 1975a. Farmer-oriented research aimed at crop intensi-
: fiction. Asian Cropping Systems Workshop, ;he International
Rice Research Institute, Los Banos, Philippines /in press/.

Harwood, R.R. 1975b.'The application of science and technology to
long range solutions: multiple cropping potentials. Conference
on Nutrition &nd Agricultural and Economic Development in the
Tropics. INCAP /in press/.

Heady, E.O., and R.C. Agrawal. 1970. Prospects and problems in
: multiplee cropping. page 82-96. In Report of the National
Seminar on Multiple Cropping, New Delhi.

Herrera, W.A.T., and R.R. Harwood. 1975. Agronomic results in an
intensive upland rice area in Batangas, IRRI Saturday seminar
series /unpublished/ International Rice Research Institute,
Los Bafos, Philippinec.

- 28 -

Ho, R. 1962. Mixed farming and multiple cropping in Malaya. Trop.
Geo. 16:1-17.

Hopper, W.D. 1975. To conquer hunger: Opportunity and potential
will. John A. Hanna International Development Lecture Series,
Michigan State-University, East Lansing, Michigan.

IRRI /International Rice Research Institute/. 1974. IRRI Annual
report for 1973. Los Bafos, Philippines.

IRRI International Rice Research Institute/. IRRI Annual report
for.1974. Los Baflos, Philippines. f/in press/.

Iso, E. 1954. Rice and crops in its subtropical zones.
Japan FAO Association. Tokyo. 611 p.

Johnson, A.W. 1972. Individuality and experimentation in traditional
agriculture. Human Ecology 1(2): 149-159.

Lee, C.S., C.Y. Lee, and C.H. Tseng. 1973. Impact of multiple
cropping diversification on farm income in Taiwan. Seminar
on Multiple Crop Diversification in Taiwan and Its Relevance
to Southeast Asian Countries. Oct. 29-Nov. 4, 1973. Taipei,

Litsinger, J.A., and K. Moody. 1975. Pest management in multiple
cropping. American Society of Agronomy symposium on multiple

Mahu, S. 1975. Maximizing utilization of rice areas in the Chiang
Mai Valley. Cropping Systems Workshop, The International Rice
Research Institute, March 18-20, 1975, Los Bafos, Philippines.

Mears,- t.'A. 1974. :Rice economy'6f the Philippines. University of
the Philippines Press. Quezon City.

Menegay, M.R. 1975. "Socio-economic factors affecting cropping systems
for selected Taiwan farmers." Cropping systems workshop. The
International Rice Research Institute. March 18-20, 1975.
Los Baflos, Philippines.

Moermana, M. 1969. Agricultural change and peasant choice in a Thai
village. University of Calif. Press, Berkeley and Los Angeles.
227 p. '

Mosher, A.T. 1969. Creating a progressive rural structure". A.gric.
Dev. Council, Inc.,'New York.

- 29--

Nair, P.K.R., and A. Singh.. 1971. Production potential, economic
'" possibilities and input requirements of five high-intensity
rotations with rice (Oryza sativa L.). Indian Jour. of Agric.
Sci. 41:807.

:Nair :P.K.R., A. Singh, and S.C. Modgai. 1973. Cropping patterns
: '' involving rice and their management. The Indian Jour. of
Agr. Sci. Vol. 43(1):70.

Norman, D.W. 1968.

Norman, .D.W. 1970.
Agric. Soc.

Why practice intercropping. Samaru Agric. News.

.Initiating change in traditional agriculture.
of'Nigeria (7):6-14.

Obradovich, M.M. 197'. i-' ehe'ral climatic map of Southeast Asia.
IRRI Library Map 188. International Rice Research Institute
Los Banos, Philippines unpublished/

Ocampo, N.R. de., E.V. Gaon, and E;M. Albano. 19 The effects of
cropping patterns on farm earning capacity in Malvar, Batangas.
Philipp. Agric. 53(1):17-27. The. University of Philippines
at Los Banos.

Oshima, Harry T.
summary and
; Countries.

1973. Multiple cropping in
further research. Seminar
in Taiwan aBd 'ts Relevance

Asian development:
on Multiple Crop Diver-
to Southeast Asian

Panabokke, C.R., and A. Walagama. 1974. The application of rainfall
confidence limits to crop water requirements in dry zone
agriculture in Sri Lanka. J. Natn. Sci. Coun.
Sri Lanka 2(2):95-113.

Philippson, W.R. 1972. Philippine crop occurrence according to
Coronas climate types: preliminary results. Phil. Agric.

Price, E.C., Jr.

1973. Aggregate supply of bovines in
Unpublished Ph.D. Dissertation, Dept.
University of Kentucky, Lexington.

of Agricultural

Rao, M.V. 1975. Cropping systems in Southern India -- Problems and
Prospects. Cropping Systems Workshop, The International Rice
Research Institute. Los Banos, Philippines. March 18-20,

Reed, R.R. 1965. Swidden in Southeast Asia. Lipunan 1(1):24-52.

Spedding, C.R.W. 1971. Agricultural ecosystems. Outlook on
Agriculture 6(6):242-247.

- 30 -

Spencer, J.E. 1966. Shifting cultivation in Southeastern Asia.
Calif. Press. 247 p.


Samson, B.T., W.A.T. Herrera, and R.R. Harwood. 1975. Draft animals
in an intensively farmed upland rice area in the Philippipes.
tRRI Saturday Seminar series /unpublished/ International Rice
Research Institute, Los Batos, Philippines.

Somnuk, S., and K. Manowalailao. 1973. Agro-economic zones and
S 'agricultural development planning. Agric. Econ.' es. Bull. No. 65
B.E. 2515.

Sritunya, S. 1975. The intensive ditch and dike method of vegetable
production in Thailand. M.S. Thesis. Central'Luzon State
University, Mufoz, Nueva Ecija, Philippines.

Thode A.R., and W. Aree. 1974. The Chiang Mai central crop
market: structure conduct and performance. Fac. of Agric.
Chiang Mai University, Chiang Mai Thailand.

Upasena, S.E., and G.W.E. Fernando. 1973. The .cropping patterns
: in Sri Lanka /unpublished/ Paper presented at the Seminar on
the Economic and Social Consequences of the Improved Seeds
at Hotel Swisse, Kandy. .

IRRI LInternatnai l Rice Research.Institute/. World Rice Statistics.
/Undated/. Dept. of Agric. Econ., The International Rice
Research Institute, Los Baflos, Philippines.


- : :, i, i

Table 1. Farm size distribution in South, Southeast, -and East Asia.a

Distribution (%)
Total less than 0.5 0.5-1.0 1.0-2-.0 2.0-3.0.... 3.0-4.0 4.0 and above
Location (000 ha) (ha) (ha) (ha) (ha) (ha) (ha)

South Asa1 64115 22 19 22 12 7 18
India 49874 21 : 19 22 12. 7 18
Bangladesh 6139 28 24 26 11 4
Nepal 2076 30 14 15 8 24' '
Pakistan 4860 17 15 17 12 8 3i- ''
Sri'Lanka 1166 39 26 19 6 6 4

Southeast Asia 19899 33 20 21 10 5 11
Indonesia : 12237 44 26 18 6 2 4
Malaysia (West) 450 15 22 31 16 6 10
Philippines 2166 %4 : 7 30 .21 12 26
Thailand 3174, 10 7 22 17 11 32'
Vietnam 1872.:: 37 19 22 10 4 8

East Asia 8405 38 31 25 6
Ja~pn 5176 38 31 25 6
Korea 2452 36 32 26 6
Taiwan 777 37 .; 29 24 10.

a/Data were taken from "World Rice Statistics," compiled by Dept. of Agric. Economics, IRRI, Los Banos,
laguna Philippines from the following sources:
: Abstract of Statistics Agriculture, Forestry and Fisheries, Japan 1970..
-.Agricultural Yearbook, Republic of Korea, 1973,
Statistical Yearbook, Thailand No. 29, 1970-1971.
Indian Agriculture in Brief, llth ediction, 19,71.
Nepal Ministry of Economic Planning, Physic4l input-output characteristics "of cereal gain proddction,... .-
for selected agricultural areas in Nepal, 1965-1966.
FAO report on the 1960 World Cencus of Agriculture, 1966.
-"2 hectares and above


16 98

cm size, portion of labor Table 2. Arable land, fagriculin agilometer of arable land.a'.ture, people per square k

People (no./eq kt Portion of labor force
)60. 1970 Fax -e m si2 artale land) in agriculture
) ha) (000 ha) : Location (00( (ha) 1960 1970 1960 1970




- woria -iace OiaLrnLI-i
Lrom the following so
books, 1962, 1981, 1I
960 World Census of
1 Yearbook, 1970-71.
arbook 1974.
:ambodia 1963.
that 40 percent of 1

3.6 1 Philippines )
3.5 ( Thailand )
1.3 ( Vietnam )

2.0 1 uKhmer Republic )
5.7 1. Malaysia (Sabah, .
Sarawak, West
;s, =.'Data were taKen irom L=u
urces Laguna, Philippines J
972. FAO Production year
\grici FAO Report on the le
Thailand Statistica
NEDA Statistical Ye
Area Handbook for C
the f. b/1960 survey indicatedwere
than 8 ha.







A. .. La.'.- AL S..tJ~ ,cItuLf.t


arms + percent were more

less than 2 ha and only 14

1.1 East Asia
1.2 ( Japan
0.8 ( Korea
1.3 1: Taiwan



10 aa

192318 122.5 'U. S. A.

199950 .

2.4 South Asia
2.5 ( India
1.4 ( Bangladesh
1.2 ( Nepal
3.5 ( Pakistan
1.6 ( .'Sri Lanka

1.8 Southeast Asia
: Burma
1.1 ( TrnA,-nciO 4


S 14!








70 -














--P A rw T? -r ---

j;U~I:!~L:'9 'L~"LI YVY L'CliLVil)

..... 18)

99 107


. rI I t-- Ir-_ >n T Tnc n t ,Z

<-*- .* -f -- t 1 _

I Ir - I TI

Table 3. Cropping intensity as influenced by size of farm in

.. .. -' Bangladesh.

Size of farm -- Cropping intensity

A. Small (under 2.5 acres) 167

B. Medium (2.5 to 12.5 acres) 148

C. Large (12,.5 to 40 acres) 130

D. Farm over 40 acres 117

i ---------i- *, --------- 'J

Source: Amed, K. 1965. Agriculture in East Pakistan.

M/S Ahmed Brothers Publication. 2 J.K. Mandir Rd.,


. ', .

Table 4. Numbers of agricultural researchers and extension workers in

various world regions in 1965 and 1974..

Agricultural Agricultural
Region researchers (no.) extension workers (no.)
1965 1974 1965 1974

Western Europe 10,187 14,752 21,291 27,382

Eastern Europe and
USSR 35,152 52,256 -

North America and
Oceania 10,082 12,939 14,136 15,764

Latin America 2,337 4,738 5,088 19,302

Africa 3,035 4,808 46,000 75,500

Asia 22,927 38,359 119,400 165,000

South 1,908 2,915 66,500 82,000
Southeast .1,150 2,164- 20,000 40,000
East 10,765 15,213 16,900 20,000
West 1,104 2,067 16,000 23,000
China 8,000 16,000 -

Source: James K. Boyce and Robert E. Evenson. 1975.
Agricultural Research and Extension Systems. Dept.
of Agric. Economics, Univ. of the Philippines,
Los Bafos, and the Agricultural Development Council,

Table 5. Rainfall Classification for Southeast Asia.

Zone I : less than 3 consecutive wet months.
(No rainfed rice crop possible)
II ;' more. than 6:.consecutive dry months.
12 .t 5 -6 ,consecutive dry months.
:t .' ';2- 4 c6osecutive dr,y months.
4 less thn'i2 co'd'secutive 'dry months.-

Zone II : 3-- w4 consecutive wet months.
.:. 'i":-:.. more than 6 consecutive dry months.
-'-(always "sharp end")
II.: i':5: -' 6 consecutive dry :months.
(with "sharp end")
S 112:" :' 5 6 consecutive dry months.
(with "gradual end!!-
S I1. 2' 4 consecutive dr.y months. -.:
(with "sharp end")
113.2 2 :- 4 consecutive dry months...;..
(with :"gradUal- end")
I' 14'i:- : less than 2 cdonsecutive dry months.:.
("bimodaPil') "
114.2 : less than 2-c':onseiutive'dry months.

SZoned il :" 5' 6" c6ns6&itivd w-t -months.-:
IIIi : more than 6 consecutive dry months. .
1112.1: 5 6 consecutive-dry months.'
(always "sharp end") ::
1113.1: 2 4 consecutive dry months.;
(with "sharp end") :'': :;
III3.2: 2 4 consecutive dry months.
(with "gradual end") .
1114.1: less than 2 consecutive dry months.
II14.2: less" thfin 2'conebutive dry months.'!:
S(". unimrodal)' .

Zone IV : 7 8 consecutive wet months.
IV and .
I2 not applicrible" .
IV3.i : 2 4 consecutive dry 'ianthts,* ..
(with "sharp end")
IV3.2 : 2 4 consecutive dry months.
(with "gradual end")
IV4.I : less than 2 consecutive dry months.
IV4.2 : less than 2 consecutive dry months.

Zone V

more than 8 consecutive wet months.
with 2 or more dry months.
less than 2 dry months.

" .'. '

Table 5 (cont'd)


"wet month" a month with at least 200 mm rainfall.
"dry month" a month with less than 100 mm rainfall.
"sharp end" there are less than 2.months between the wet and the
dry period, that have 100-200 mm rainfall.
"gradual end" there are at least 2 months between the wet and the
dry period, that have.100-200 mm rainfall.
bimodall" -.a rainfall profile with a secondary rainfall peak of
more than 200 mm rainfall for at least one month.
"unimodal" a rainfall profile without a secondary rainfall peak
of more than 200 mm rainfall.
Zone I no rice cultivation possible without additional water.
Zone II only one early maturing variety may be possible. Rainfed
paddy risky and not all.area will be planted;except in a wet
Zone III sufficient time for one late variety. An early variety
plus additional upland crop also possible. Two rice crops
possible if the first is direct-seeded and early-maturing.
Zone IV 2 subsequent rice crops possible. Multiple cropping with
one rice crop definitely possible.
Zone V continuous rice cropping possible.

The primary subdivisions are chosen on the length.of the dry

sub 1 more than 6 dry months...
sub 2 5-6 dry months. .
sub 3 2-4 dry months.
sub 4 less than 2 dry months.

This reflects the period when not enough water is available
for any crop.

Each subdivision can then be characterized by the, .type of
rainfall profile: unimodal or bimodal and by the type of transi-
tion period: a sharp transition or a gradual transition.

sub-sub (a) A secondary subdivision (a) indicates that at least
1 month has more than 500 mm/month, making conditions undesir-
able for upland crops unless drainage is excellent-.

Table 6. Percentages of land in Bangladesh, Indonesia, and the
that fall under 5 major rainfall categories.


Rainfall category Bangladesh Indonesia Philippines

I less than 3 wet monthss 25 12
II 3 4 wet months 23 40 28
III 5 6 wet months 72 30 37
IV 7 8 wet months 5 5 21
V more than 8 wet months 2

Table 7. Percentages of arable land in rice and the percentage of land
irrigated in 1970.

Arable land Irrigated land
in rice % of arable
(%) (%)

South Asia 25 21
India 23 17
Bangladesh 82 6
Nepal 61 9
Pakistan 8 65
Sri Lanka 31 23

Southeast Asia 43 19
Burma 45 4
Indonesia 45 38
Khmer Republic 80 3
Laos 70 2
Malaysia (Sabah, Sarawak, West Malaysia) 20 7
Philippines 35 9
Thailand 59 28
Vietnam 86 20

East Asia 56 47
Japan 53 51
Korea 52 53
Taiwan 90 58

China 30 69

U. S. A. 0.4 8


FAO Production yearbooks 1962, 1971, 1972.
FAO Report on the 1960 World Census of agriculture.
Thailand Statistical Yearbook, 1970-71.
NEDA Statistical Yearbook 1974.
Area handbook for Crbodia 1963.

Income from crops ( per hectare)

Units of management per hectare

Fig. 1. Relation of farm income to
and temperate regimes.

management in tropical

"Community" owned
production resources


Farm resources


Marketed goods

and services

Purchased goods
and services

Figure 2. Conceptual model of a Nepali hill farm production system. (from Ilarwood. 1974).

--- for months w/ rainfall greater than 200 mm/mo.
.--.. for months w/ rainfall less than 50 mm/mo.
with mechanical power only.

Primary tillage capability class
2 5 4

Soil Textural Class

(2:1 clay type)

(1:1 clay type)





Percent Increase in bulk density when puddled

4.1 -8

8.1 12


Crop Potential After Paddy Rice (Subject to Water Availability)

-- grows well
---- grows with

Rice (transplanted)
Vine crops

Figure 3 Soil classification categories which indicate multiple cropping potential
in puddled rice soils having limited water (March 1975 estimate).






Form productivity

Quantity of inputs

Figure 4. Comparative efficiencies of different farming systems
with respect to use of inputs.

Figure 5. Asian cropping systems network present and planned sites.


y deK 4Khon Kaen
Joydepur THAiLAND P. Mai
Rangoon- Pangasinan
Chieng hainat Ubon Batangas
Mai Bangkok/ S VIETNAM ",
Bangkok cCAMBO-

Meglleww aha Illuppallama Ku
Megallewawa Kuala
SRI LANKA ^^ Trengganu

Central Lampung- *
Sukamandin~ "
* Outreach site Indramayu
* Collaborator


Phase of

(in produc-
tion areas)




No Farmer Involvement Necessary Farmer Involvement Necessary

Of Technical and Support Factors Of Socio-Economic Factors

Physical Biological Support systems Resource quantities Social attitudes

water availability varieties extension flows and relation- traditions
temperature diseases credit ships (land, labor, goals
sunshine insect input distribution capital, management needs
soil weed control price
government policy



I \

Improved I Feedback on needs
component for attainment of
technology j potential



Feedback on \ feedback n
changes needed changes needed

SA. Test for physical and
biological adaptation /

B. Test for compatibility
with farming systems


i/As oiutlined b)y participants of the 1975 Cropping Systems Workshop.