a flow rate of 500 ml/min. This equipment gave droplet sizes averaging about 20 im with the cone jets. Experimental Design: A 4 x 4 latin square design was set up at the Graeme Hall Research Station in Christ church, Barbados. The treatments within the latin square were: 1. Oil treatment of inoculated plot 2. Oil treatment of uninoculated plot 3. No oil treatment of inoculated plot 4. No oil treatment of uninoculated plot The plants used in the experiment were zucchini squash, the size of each plot within the latin square was 25 m2 (5m x 5 m) with a population of eighty-one zucchini squash plants. Inoculation: The plots were inoculated by planting in the middle of plots (a) and (c), one squash plant known to be carry- ing symptoms of mosaic virus. Inoculation took place five days after crop emergence. Application of Mineral Oil: The oil was applied initially at five days after seedling emergence (at that time over 85 % of the seeds would have germinated) and then two times weekly throughout the period of rapid growth. Subsequently the appli- cation of oil was reduced to once per week and continued through harvest. Between March 18 and May 14, 11 applications were made. The oil was applied at the rate of 17.6 I/ha (2 gal/acre) or 0.75 % emulsion in tank mixes. A suitable fungicide was appliced on a weekly basis to control mildew diseases. There was con- tinuous harvest for five weeks. Disease readings were made each week on the plants from all replicates beginning about four weeks after emergence. At the end of the five-week harvest, stand counts were taken along with counts of diseased plants per plot. RESULTS The viruses identified in the cucurbits from each of the islands surveyed are shown in Table 1. There was evidence of complex in- TABLE 1. Occurrence of virus diseases on cucurbits in the West Indies 1979-1984 West Indian Cucurbits Cultivated Island Water Surveyed Squash Pumpkin Cucumber Cantaloupe Melon Antigua A, C A, D A, D Barbados A,B,C A A D A, D )Viruses Grenada A A A, D A, D )Identified Montserrat A, C A,B,C,D A B A, D Jamaica A, C A,B,D A, C A, D A, D ) St. Lucia B, D A A, D A A A CMV; B = SMV; C = SMV/CMV Complex; D = Water Melon Mosaic E = Beta virus 1; = Crop not grown at time of survey fections of squash and pumpkin by CMV and SMV in samples obtained from Jamaica, Barbados and Montserrat. None of the antigens prepared from the cucurbits reacted serologically to beta virus 1. Profuse growth of french weeds (Commelina elegans) car- rying CMV was found within and without some curcurbit plots in Montserrat, Jamaica and Antigua. Mosaic symptoms were present in the unsprayed uninoculated plots within four weeks after crop emergence. After seven weeks, 10% of the plants showed symptoms. In the unsprayed inocu- lated plots 25% of the plants showed symptoms, while in the inoculated plots that were sprayed 3.7% of plants carried symp- toms and 5 % of the sprayed uninoculated plots carried symptoms (Table 2). Overall plant appearance and disease control were better in plots treated with oil (P = 0.05). The oil spray gave a better con- -trol in plots that were uninoculated, but the differences between inoculated and uninoculated were not significant (P > 0.05) (Table 2). The interaction between oil treatment and inoculation did not significantly affect disease spread (P = 0.05). Unsprayed plots, in- oculated or not, displayed very severe symptoms with widespread stunting and partially killed plants; fruits were mottled and twisted and in some cases highly irregular in shape. DISCUSSION Complex infections of cucurbits by two or more types of viruses seem to inflict a more severe type of symptom. Complex infection of cucurbits by viruses was first reported from the West Indies in 1979 from Jamaica (Turner and Stace-Smith, 1979). Montserrat's report of complex infection highlighted severe symptom develop- ment resulting in total crop loss (Thomas, 1980). Vegetable growers in the West Indies have long been plagued by large numbers of aphid-transmitted virus diseases for which only limited control measures have been available. Virus disease control in these crops is aimed at the widespread use of insecticide and growing resistant varieties. This control measure has not proven very successful in these islands. In Barbados one field of squash recorded an occurrence of 45% mosaic within 6 weeks of crop emergence, in spite of well-timed applications of Ambush and Diazinon insecticides. All the cucurbits cultivated in Jamaica, Antigua and Grenada were afflicted by CMV (Table 1). The squash which was used to test the efficacy of the mineral oil suffers natural epiphytotic levels in all the islands in which it was cultivated. Being the only cucurbit af- flicted by all viruses (Table 1), it was used for the disease control test. From Table 2 it is evident that whether or not a field was ar- tifically inoculated, the disease will spread if viruliferous aphids are in flight. This means that disease spread is not necessarily con- tingent to the presence of in-field inoculum. From this experiment there was a 2.5-fold increase in disease spread in inoculated versus TABLE 2. Disease development (within treatments) per plot Di sease Plot No. of Total No. Mean No. of Disease Replicates of Plants Plants per Replicate Total No. of Cumulative % % % Plants % Severe Moderate Mild (A) Oil 4 324 3 13 3.7 0.0 1.1 2.6 (B) No Oil 4 324 20 81 25.0 11.2 6.8 7.0 (C) Oil 4 324 4 16 5.0 1.0 0.5 3.5 (D) No Oil 4 324 8 33 10.0 3.0 6.0 1.0 A = Inoculated and Sprayed with Oil B = Inoculated and Unsprayed C = Uninoculated and Sprayed with Oil D = Uninoculated and Unsprayed PROCEEDINGS of the CARIBBEAN FOOD CROPS SOCIETY-VOL. XX 294