TABLE 5. Results of on station observations on introduced yam cultivars.


 Variation in tuber yield per plant could be attributed to the in-
 tensity of anthracnose in White Lisbon yam. However, genetic
 variations could have contributed to the lower yields of V 1712
 (D. cayenensis) and SEA 189. Anthracnose levels of 0.8 and 0.0
 were recorded for these respectively. The highest anthracnose
 level was evident in Belep. Ninety-five percent emergence was
 recorded in all cultivars except V 1712 which gave only 10%.
 Since this cultivar is D. cayenensis, only a larger sett size from
 head and tail portions of the tuber should be used as planting
 material.
 Negligible insect damage was noticed on the foliage of all
 plants. Though some boring insects were observed in the tubers
 at harvest, these resulted in superficial damage only.
 The foliage of all-plants was examined for virus symptoms.
 Mottling occurred in a few plants of Belep and these were im-
 mediately discarded. Symptoms were not observed on other
 plants. When 2% of all harvested tubers of each cultivar were
 sliced and examined, internal brown spots were noted in a single
 tuber of Belep. Table 4 shows that only the planting material
 from CARDI, Barbados had been virus-tested.
 Data for 1983 showed increased levels of anthracnose except for
Belep which remained the same, 2.0. These increases ranged
from 2.5 in White Lisbon to 1.2 in V 1712 grown at La Plaine.
Similar results were recorded in all those locations in which White
Lisbon was grown. This is associated with the similar climatic
characteristics. Generally, depression in yield was observed in
White Lisbon subjected to the virus-free process indicated in data
recorded in 1982. Apparently Belep and SEA 189 can tolerate
higher levels of anthracnose than White Lisbon.
 The expression of anthracnose in virus-tested White Lisbon
with increasing generations has also been reported in St. Lucia,
St. Vincent, and Grenada. In all cases the levels noted were
greater than those observed in Barbados which has a much drier
environment, so that results reported herein may be attributed to
either the greater precipitation in Dominica or the virus-tested
process predisposing the planting material to more pronounced
expression of anthracnose. Further work is necessary to elucidate
the incidence of anthracnose in the introduced material in
Dominica, and to identify appropriate technology to control this
disease. Thus it may be more appropriate at this stage to accept
the lower yields of SEA 189 concomittant with virus-testing the
same, rather than persisting with White Lisbon. Nevertheless, in
the interim, two additional cultivars, Coconut Lisbon and Orien-


tal, both virus-tested and D. alata, have been brought in to
Dominica and planted in 1984. Preliminary observations indicate
that Oriental, grown at the Corona Station (precipitation 6,250
to 7,500), shows very little anthracnose incidence. This confirms
observations in Barbados.
 Differences in approach between 1983 and 1984 are attributed
to the adoption of CARDI's Farming Systems Research and
Development Methodology, in which emphasis is placed not only
on technology generation but technology transfer. The farmer
will maintain much closer contact with the researcher through
participation with extension agents and the researchers on the
farm. Closer examination of the farmer's resources, objectives and
decision making processes is achieved by the FTC who reside in a
rural area. This facilitates evaluation of the impact of the new
technology. In addition to evaluation in 1984, generation of new
technologies has been initiated. A yam collection of 22 local ac-
cessions has commenced on a farm in La Plaine in order to ascer-
tain the characteristics of indigenous material. Well-adapted
material with disease tolerance could be identified and virus-
tested at the CARDI, Barbados facility for introduction into the
farming system.
 The association of high yam yields with pen manure, alluded
to earlier, will lead to on-farm tests on this topic, since pen
manure is present on most small farms.







Acknowledgements
 CARDI and FTC Dominica wish to thank those farmers, extension agents and
staff who made this work possible. The contributions of MOA were appreciated and
the support of the management of CARDI/USAID FSR/D Project was essential for
completion of the above.
References
 1. Fergusson, T.U. 1981. Root crop production of project farms in Dominica.
Consultant report #11, CARDI.
 2. Henderson, T.H., and P.I. Gomes. 1979. A profile of small farming in St.
Vincent, Dominica and St. Lucia. Report of a Baseline Survey for CARDI, Dept.
Agricultural Extension, U.W.I., St. Augustine, Trinidad.
 3. Thomas, M. 1984. A market analysis of aroids as compared with other root
crops in Dominica. Report CARDI/EDF Aroids/Arrowroot Project, Botanic
Gardens, Roseau, Dominica.


PROCEEDINGS of the CARIBBEAN FOOD CROPS SOCIETY-VOL. XX


Location Year Species/Cultivar No. of Anthracnose Insect damage Virus symptoms Virus symptoms Average Yield/
 Plants damage (0-5) (0-5) leaf (0-5) tuber (0-5) plant (kg)

La Plaine 1981 W. Lisbon (A) 50 data not collected 1.6

La Plaine 1982 W. Lisbon (B) 238 0.7 0 0 0 4.8
La Plaine 1982 W. Lisbon (C) 197 1.2 0 0 0 2.8
La Plaine 1982 Belep 35 2.0 0 0.2 0 1.4
La Plaine 1982 SEA 189 23 0 0 0 0 1.7
La Plaine 1982 V 1712 28 0.8 0 0 0 0.5

La Plaine 1983 W. Lisbon 483 3.5 0 0 0 1.0
La Plaine 1983 Belep 180 2.0 0 0 0 1.2
La Plaine 1983 SEA 189 80 2.0 0 0 0 0.5
La Plaine 1983 V 1712 14 2.0 0 0 0 1.1
Grand Bay 1983 W. Lisbon 264 3.0 0 0 0 1.1
Portsmouth 1983 W. Lisbon 179 2.0 0 0 0 1.4


258