in definition made it impossible to say how much of the increase is actually in small plots
and how much in farms of less than five acres. In particular, the Trinidad and Tobago fig-
ures for small plots in 1956 is probably inflated by omissions from other size groups.

 If comparable figures for the Leeward Islands were available we should expect to find
an increase in small farms at the expense of estates, because estates have been broken up
in Nevis and Montserrat to make land available to small farmers. In Antigua, a similar
government scheme has resulted in allocation of estate land to peasants, but the big move-
ment took place in Antigua in the 1930's when considerable estate acreages were allocated
to peasants. In British Guiana, we should expect to find in recent years an increase in the
relative importance of small farms. This is not because of the breakup of estates so much
as because of new acreages reclaimed by drainage schemes being generally made available
to peasant farmers.

 Provisional figures for the 1961 census give some indications that migration from the
country into the town, in Jamaica, and migration outside the territories themselves may
have reduced the number of farms, particularly those of less than 10 acres. Until figures
of acreages are available, however, it is difficult to say whether this trend has reduced
acreages in each size group. It may mean that some operators are farming slightly lar-
ger areas, which may affect the total acreage within size groups only marginally. There
is, however, some evidence that emigration has reduced the cultivation of roots and cer-
tain vegetables. Available information shows this as the only sector for which we can iden-
tify a decline in production totals.

 Although figures of sizes of farms growing certain main crops were estimated in va-
rious censuses these are not felt to be reliable enough to permit comparisons between
dates. For certain crops, figures of deliveries from large and small farmers are avail-
able, and we have been able to assess the changing position from these.

 Sugar is grown by peasants and estates in Jamaica, Barbados, Antigua and Trinidad
and Tobago, but in St. Kitts and British Guiana, peasant cultivation is negligible and the
peasants' share in the total has not shown any change in the past two decades.

 In Jamaica, the peasants' share of total cane supplied to mills increased from 25.5% of
the total in 1940 to 38.8% in 1955, and to 44.2', in 1959. We believe that this increase in
the peasants' share has been one of the main factors, if not the main factor, in depressing
yields in Jamaica when they have been rising in most other territories. We do not however
envisage a continuation of this trend towards greater peasant participation in cane produc-
tion.

 In Barbados it is estimated that peasants' acreage reaped increased from about 7,000
or 18% of the total in 1940-46 to 10,000 or about 22% of the acreage in 1955-60. In Antigua
peasant acreages are estimated at about 40% of total acreage, and there has been little
change since 1950. In all cases, peasant yields are slightly lower than those of estates,
so that a 40% share of the acreage would indicate less than a 40% share in actual cane de-
liveries.

 Table 13b shows the size distribution of banana growers in certain territories. No fig-
 ures are available that would show if the size pattern of banana farms has changed. We