in production. Some land may be made accessible by new roads. Where this is impossible
however, we do not consider that, within the income structure we have postulated, farmers
will increase cultivation of inaccessible patches of soil in the mountains in Jamaica or in
the Windward Islands, such as has happened in Grenada's recent past.
Projected land use pattern
 Table 9a shows the present land cultivation pattern as observed at the most recent cen-
sus dates. In Tables 9b and 9d these areas have been projected on the following assump-
tions.

 The total acreage of crop land is derived from aggregating the average projections made
for the individual crops, which is explained in the next chapter.

 The acreage under pasture is derived also in relation to projections of livestock num-
bers, although only for improved pasture can these be considered a fairly exact relation-
ship.

 In Jamaica, we have assumed some increase in the land in farms between 1958 and 1965.
 On the other hand, it is suspected that, due to migration, marginal land may be going out
 of cultivation; we do not predict much if any increase in total land in farms between 1965
 and 1975. We do, however, consider there will be soma increase in the ratio of pasture
 to woodland and an even bigger increase in the proportion of pasture that is improved.

 Since so much land in Trinidad and Tobago is tied up in oil leases and crown forests, it
 is not unreasonable to suppose some increase in the area of farmland in the period. We
 consider, however, that after 1970 any increase in the area of farms is likely to be asso-
 ciated with increased pasture rather than crop land. The increases are not expected to be
 large.

 For the Leeward Islands, Windward Islands, and Barbados, we have predicted small in-
 creases in farm areas since there is some room for increased farm use of land in Domi-
 nica. If the hoped-for expansion in bananas takes place, it is expected that new areas will
 come into use, although a considerable part of the expanded acreage would probably come
 from land already in farms but not cultivated. Although we expect some increases in pas-
 ture land improvement, it appears that in the small islands the biggest part of the increase
 will be in crop land.

 In British Guiana we have assumed a rather greater increase in the farm area. This
 follows from known plans to increase rice acreage and to bring unoccupied land into rice
 cultivation. We expect also that cattle ranching may spread to some unoccupied land, thus
 increasing the area of farms further, although improved grasses may lead to intensive rath-
 er than extensive expansion of this industry. The assumptions mainly based on our pro-
 jections for rice, sugar and cattle cultivation indicate an increase of farm land from 5% to
 T7' of total land area, and the proportion of crop land to total farm land isassumedtoin-
 crease from 11% to 14%.

 We should like to think that future increases in production will come from a better uti- v
 lization of land rather than bringing of more marginal land into cultivation. Already one