The African striped ground squirrel, Xerus erythropus (E. Geoffroy), has been found to constitute a serious pest to maize seed at the planting stage, causing mean losses of 9.7% and accounting for 57.3% of total damage found. A feature of ground squirrel damage is its unpredictable nature. Methods of reducing looses of planted maize seed to X. erythropus at the subsistence farmer level in southern Kenya were investigated. Constraints affecting a control programme by farmers were identified as follows: low standards of living and education, limited financial resources, strong individualistic attitude of farmers and small field size in relation to the home range size of squirrels. Removal trapping and poison baiting were selected for trial as meeting requirements of ease and simplicity, and involving materials available to farmers at the time. Field trials of bromadiolone and difenacoum anticoagulant rodenticides and removal trapping over periods of 1 month and 3 months prior to the expected onset of the rains failed to affect damage levels significantly on individual field units. Underbaiting and a high reinfestation rate were considered to be the primary causes of failure, and the habit of scatterhoarding exhibited by X. erythropus further complicates the poison baiting trials. The unpredictable nature of ground squirrel attack discourages farmers from expending valuable resources on control. Alternative strategies for farmers are discussed.

St. Helena is an Atlantic Ocean island lying 1,200 miles off the coast of Angola. Both species of commensal Rattus occur on the island, the ship rat (R. rattus) having possibly arrived as early as 1502, the year of the island’s discovery, and the brown rat (R. norvegicus) in the 1700s. Today, rats are widespread and common over the entire island, including the arid wastes, the inhabited and agricultural areas, and into the National Park around the central peaks. The impact of invasive rats on the endemic flora and fauna (predominantly invertebrate with many endemic species of snail, spider, and weevil being recorded) is unknown. The only natural enemies of rats on the island are occasional feral cats, raptors, and other predators being absent. In 1924, the Agriculture and Forestry Department started a 50 year poisoning campaign against rats which ended in 1956 when the brown rat was considered to be almost extinct. Since that time a continuous island-wide suppression campaign has been carried out by the Department of Public Health. This consists of placing anticoagulant baits in and around the inhabited areas of the island (and, on demand, in the agricultural and forested land) routinely, checking baits at least every two weeks and more often if take is high. On average, 20 tons of bait are laid every year. Data on baits placed, takes, and dead rats found were examined for the years 1995 to 1998. R. norvegicus was found to be the most common species, with the two species approaching 1:1 in only three areas of the island. A number of recommendations for the improvement of the rodent control program have been made with the overall aim of raising the professional status of the operators from that of "rat baiters" to that of professional pest control technicians. This will effectively raise the professional status of the Environmental Health Section as a whole. The aim is also to involve the public as partners for improvements in their own health and welfare.

Introduced commensal rodents have had a major impact on the biota of island communities worldwide; the ship rat (Rattus rattus) and the house mouse (Mus domesticus) have a long history in the Galapagos islands, while the larger, more aggressive brown rat (Rattus norvegicus) was identified only in 1983 on one island, Santa Cruz. By 1988 it had spread into the agricultural zone but was still restricted to human habitation. In 1993 the cross-island road and village communities in the agricultural zone of Santa Cruz were sampled using a standard trap line of break-back traps. House surveys were also carried out, where appropriate. The brown rat was found to occur at all sites sampled on the south side of the island, including sites independent of human habitations, and is now the dominant commensal rat in houses. The impact of the brown rat on the ship rat and the conservation implications of its spread are discussed.

The Secretariat for Rural Development of the State Government of Chiapas, southern Mexico, has been charged with conducting a campaign against the field rat which began in February 1991. Four areas were identified as key sites for the campaign benefiting 3,355 farmers in a total area of 8,000 ha and an initial budget of approximately $500m Mexican pesos (US$170,000). The major crops for which damage is reported are maize, sugar cane and cacao. The term "field rat" is a general one encompassing any rodent causing damage to field crops and in Chiapas probably covers a variety of species; no studies have been done to identify the species or quantify losses to crops. Personnel had no previous experience in rodent control and no resources to permit preliminary investigations in the field. A method for assessing field damage levels was developed and fields were treated with zinc phosphide (high damage), diphacinone (medium damage) or untreated (low or no damage). A decrease in subsequent losses was reported by farmers involved in the campaign. The campaign in 1992 is restricted by financial and logistic constraints as the field rat campaign has been united with locust and other field pests in a single campaign entitled "Control of Pests to Basic Crops" with a much reduced budget overall. Problems found in the 1991 work and the limitations and of the campaign are discussed.

The impact of introduced commensal rodents on island flora has been relatively little studied compared with their impact on the fauna. The effects on vegetation composition, regeneration, and decomposition are largely unknown, but potentially great. Preliminary studies were carried out in the Galapagos Islands between 1993 and 1994 on the diet of introduced rats, Rattus spp. and feral house mice, Mus musculus, seed recovery rates and subsequent germination rates of seeds. R. rattus diet was primarily vegetation and 48% of rats had seeds in their stomachs. Significant differences were found between body size and overall contribution of both vegetable and animal foods, larger rats eating proportionately more animal foods and less vegetable. There was no significant difference between the sexes in terms of main dietary components. There was no significant difference in the selection of food types between R. rattus and R. norvegicus, both species tended to prefer banana and avocado, and only rats from the Miconia zones showed a preference for Miconia berries. No intact seeds were found in the stomachs of feral house mice from the same sites. Recovered seeds of two native and two introduced plant species were successfully germinated under laboratory conditions. R. norvegicus is potentially a better dispersal agent for seeds as it has a greater tendency to ingest them intact. The implications of these findings for the conservation of island flora are discussed.