The results of a serological survey of ruminant livestock in some countries of the Caribbean and South America for type-specific antibody to bluetongue virus are reported. Using the microneutralisation test with the international serotypes 1 to 22 of bluetongue virus, antibodies to several types were detected. Analysis of the data indicated that in 1981-82 bluetongue virus types 6, 14 and 17, or viruses closely related to them, were infecting ruminants in this region of the world. Antibody to the related virus of epizootic haemorrhagic disease (serotype 1) was also detected in cattle. The difficulty in interpreting the epidemiological significance of data generated by a serological survey of this kind is discussed.

A serological survey of 6250 sera from cattle, sheep and goats in seven Caribbean and two South American countries showed that antibody to bluetongue virus was widely distributed in each species throughout the survey area. Overall prevalences of antibody were 70 per cent in cattle, 67 per cent in sheep and 76 per cent in goats as assessed by an immunodiffusion test. Within countries the percentage prevalences were Jamaica 77, St Kitts/Nevis 70, Antigua 76, St Lucia 82, Barbados 61, Grenada 88, Trinidad and Tobago 79, Guyana 52 and Surinam 84. No clinical cases of bluetongue have been confirmed in the area surveyed and there are no virus isolates available to indicate which serotype(s) of virus is/are causing the infection(s).

When monitored by the agar gel immunodiffusion test for antibody to bluetongue viruses, a sentinel flock of twenty-five lambs remained seropositive through the year, whereas in a sentinel herd of twenty calves only two individuals seroconverted and these became negative again within 2 months. A light trap operated with the calf herd yielded high numbers of Culicoides insignis Lutz (over 18,000 per trap night) along with C. filariferus Hoffman, C. pusillus Lutz, C. leopoldi Ortiz, C. foxi Ortiz, C. limai Barretto, C. diabolicus Hoffman and C. guyanensis Floch and Abonnenc. Culicoides were trapped at the sheep station which had housed the lambs 3 years following the sentinel study. No virus was isolated from pools of C. insignis, C. filariferus and C. pusillus. Six other species were collected in insufficient numbers to warrant attempted virus isolations.

Following the first diagnosis of campylobacteriosis in Jamaican cattle a field study was undertaken to determine the pathogenicity of Campylobacter fetus subspecies venerealis Jam (Jamaican strain) and to evaluate the effectiveness of vaccination in controlling the disease. A total of 46 nonpregnant yearling heifers and four two-year-old bulls were used in two separate experiments. The results showed that C fetus subspecies venerealis Jam readily colonised the reproductive tract of susceptible heifers and persisted in some animals (68 per cent of unvaccinated and 33 per cent of vaccinated animals) for the duration of the experiment. Pregnancy was confirmed in 13 of 18 (72 per cent) culture-negative heifers but in only eight of 28 (29 per cent) of the heifers with two or more positive cultures. Vaccination appeared to be curative because 44 per cent of vaccinated heifers were cleared of infection whereas 85 per cent of unvaccinated, inoculated heifers remained infected for at least 17 weeks. Vaccination improved the fertility level of the infected heifers threefold. Infection was not established in vaccinated bulls used for breeding infected heifers.

Variation in the percentage of lambs seroconverting to bluetongue viruses was seen between sites and years in Barbados. Transmission at some sites was nearly absent whereas all lambs at one site became seropositive. The agar gel immunodiffusion test for bluetongue gave consistent results in series of serum samples from 112 of 121 sentinel lambs. Collections of biting midges in association with sheep yielded six species: Culicoides insignis Lutz, C. pusillus Lutz, C. phlebotomus (Williston), C. furens (Poey), C. jamaicensis Edwards and C. trilineatus Fox. The first two species comprised 92% of those caught during a sentinel lamb study and were the predominant species trapped for virus isolation. No viruses were recovered from 5517 C. insignis, 614 C. pusillus, three C. trilineatus and two C. furens placed into pools during two brief intensive trapping operations.