Figures of tree population for citrus are shown in Table 15a. Since figures for acreage!
 of both citrus and coffee are considered of dubious reliability, we have not detailed the pro
 section any more than to present estimated acreages in Tables 10a to 10d; but we have pre-
 sented detailed figures of final, total supplies for citrus (Table 15b) and coffee (Table 16).

 For cocoa, estimates have been more easily obtained, since there have been major re-
 planting schemes in all major producing territories which plan to replace all old planting
 with new clonal material. However, estimates going further back than 1958 are still of
 extremely dubious reliability. So we have not considered it relevant to include a past time
 series (Table 17a). The projections shown in Table 17b are based on the replanting pro-
 gramme. Estimates in more detailed form are shown in Tables 17c and 17d. Production
 from different aged trees, and acreage bearing at different dates can be derived from these
 tables.

 For coconuts, as with other tree crops, acreages and figures of tree population are dif-
ficult to come by. To some extent, we have had to work backwards from known data re-
garding final tree yields and land use as between non-bearing and bearing trees to derive
most of the information on scales of production so that our figures must be considered vere
provisional, particularly in respect to Trinidad and Tobago and Jamaica. There are very
definite limits to land available for coconut cultivation in the West Indies since this crop
will grow well only in fairly low lying, but well-drained land. Our projections of acreage
are thus trends towards limits. In projecting production of copra, we fitted an asymptote
curve of the type yc = k + a(bx) with known acreage as the limiting factor. This was also
a close fit with figures of copra production over the past decade. Of course, improvements
in production per acre could alter this trend, but we have felt with most crops perhaps
more particularly with this one -that the land limitation is a real one within this period.
Even in British Guiana, where a number of sand reefs could be used for coconut cultivation
the additional acreage possible is not high as most of the coastland has poor draining, un-
less artificially drained-usually not an economic proposition for coconut cultivation.

 The figures for coconut acreage and tree population are shown for a base year and for
 1965, 1970 and 1975 in Table 18b, and figures for past copra production in Table 18a.

 No accurate figures exist to show past trends in acreage of domestic food crops. So
much of these products is grown on small lots, not recorded in the 1943-46 census, that
a comparison between figures at that census and later surveys is not valid. We believe
however, that the acreage of root crops is declining in Jamaica and possibly also in other
territories, and we attribute this partly to migration, partly to urbanisation and partly to
the growth of banana and cocoa cultivation by peasants, particularly in the Windward Is-
lands. Early and very provisional figures from the 1961 census also indicate that a num-
ber of small, mainly subsistence farms may have gone out of production in Jamaica due,
no doubt, to migration, or in some cases to farmers finding employment in the mining in-
dustries.

 We expect a continued decline in the acreage of roots and little change in the acreage or
vegetables. The acreage of corn and pulses appears to have been declining in all areas
since 1950 and we expect this trend to continue. Our conclusions are summarised in
Table 19.