Dear All,

There are some very interesting aspects in the Nepal data that
potentially offer promising entry points for further research :

1) Also under the Nepal conditions it is reported that the crop
growth cycle under SRI is reduced by comparison to the conventional
system by anything between a few days up to 3 weeks. Some scientists
including myself in Cote d'
Ivoire have found the opposite - - , which can be explained in
principle quite logically by looking at the respective root systems
(under SRI still growing actively, while under flooded conditions the
roots were suberised and senescent). It seems to me that this feature
(a delay or advance in flowering) may well be critical in arriving at
a more fundamental / scientific understanding of what precisely is it
that causes the yield increases under SRI. So plenty of scope for an
interesting (PhD) research project!

2) For the Nepal plots no effects were recorded from the use of older
transplants (17 days versus 8 days). Actually the average yields were
highest for the older transplants. This is in agreement with what I
found in the Cote d'Ivoire experiments.
To me it indicates that for the "SRI technological package" the
components of "single transplants" and "wide spacing" are of
relatively greater significance in obtaining high yields than is the
component of "very young transplants". Certainly for a farmer and in
practical terms this is a potentially important result, making the
initial post-transplanting phase under SRI less delicate and easier
to manage than it is with the present recommendation.

Best regards,

Willem

Norman Uphoff wrote:

>Rajendra Uprety, agricultural extension office for Morang district
>in Nepal, has taken up SRI very systematically with farmers in his
>area. After some small demonstrations last year to satisfy himself
>and farmers that SRI really works, he got several dozen farmers to
>try the methods under supervision. His attached report gives
>details, and shows a 133% increase over the controls with standard
>methods -- from 3.77 t/ha to 7.85 t/ha average. I have shared this
>with some scientists who are skeptical/critical about SRI, and the
>response was that these yields are not possible.
>
>The data which are simplest and most impressive are those on length
>of time for the rice crop to mature. Some farmers had their crop
>mature three weeks early but the average was two weeks, something
>quite remarkable. In Andhra Pradesh, the average shortening of the
>cycle has been 7-10 days, already remarkable. In Cambodia, CEDAC
>reports 5-7 day shorter cycles. We don't know what is going on in
>the Morang soils. Some will dismiss these data too as impossible.
>But I think Rajendra was trying to be careful in all the data
>collection and reporting.
>
>There could be some faults in details, but measurement errors,
>particularly for something like data of planting and date of
>harvesting, are unlikely to eliminate the average differences
>reported. I am circulating this to encourage you in your respective
>countries to keep records on length of crop cycle so that by next
>year, we can have a very large and well-distributed data set. If SRI
>methods can give higher yields within a shorter period of time, the
>per-day production of photosynthate is increased even more. That
>should be of real scientific interest.
>
>If you have any data on crop cycle already from this past season or
>earlier ones, I would be glad to receive those data and will put
>them together in an interim report to be shared within the SRI
>network. With best wishes,
>Norman