188 Participatory Processes
is 1470 takas per farmer (n = 145).^6 This is equivalent to 9800 takas (US$204) per
hectare of rice land over the year. The inputs required are mainly seed, organic
compost for pit preparation and labour. Farmers usually just use a portion of the
fertilizer that they have purchased for their rice production on their vegetable dyke
crops and the organic compost is produced at no cost.
Vegetable seed is increasingly produced locally at no financial cost. One study
conducted in the highest external input use district showed that the number of
women participants producing their own vegetable seed went from 26 per cent to
69 per cent (n = 200 respondents from 21 FFS).^7 This helps to make the activity
reasonably remunerative.
Trees on dykes
There are cultural practices, such as periodical pruning of roots and branches,
which make it possible to grow trees on the dykes without affecting the rice crop.
This type of pruning is not favourable to fruit trees but is perfectly appropriate to
trees producing timber, cooking fuel and fodder. In December 1998, 28 per cent
of project farmers (n = 19,450 farmers) were experimenting with this technique.
We have observed that two years after the project had left, farmers continue to
tend their trees. It is too early to evaluate the economic benefits of this technique,
but we are already observing that the number of farmers planting trees on their
dykes is growing and that many have initiated small-scale tree nursery businesses
to supply their community with the required tree saplings.
Rice–fish–dyke crops
Since the integrated system does not decrease the yields of any one component,
rice farmers who integrate fish and dyke crop production naturally compound
their net returns. Already, 20 per cent of the poor and marginal farmers in the
programme have adopted this combined practice.
Overall Economic Impacts
The exact economic impact of these practices on households is difficult to assess.
The figures given above are based on a single 0.11ha plot whereas our participat-
ing households own about four plots averaging a total of 0.45ha on which it is
impossible to say how many of the innovations are practised. Another difficulty
is that households consume much of the increased production as food and donate
a significant portion to relatives and neighbours, making an economic analysis
nearly impossible. Even if the actual returns per household attributable to the
innovations were known, we still would not know how much this would repre-
sent relative to the total income of the household and its needs. Nevertheless, the