Field experiments were carried out at Coimbatore, Tamil Nadu State, India, for four consecutive years (1990–94), to monitor the seasonal activity of Cotesia flavipes Cameron (Hymenoptera: Braconidae), a larval parasitoid of gramineous borers, including the sugarcane shoot borer (Chilo infuscatellus Snellen), sugarcane internode borer (Chilo sacchariphagus indiens (Kapur) and sorghum stemborer (Chilo partellus Swinhoe) (all Lepidoptera: Crambidae). The highest levels of parasitism (0.0–17.9%) were recorded on Ch. partellus followed by Ch. sacchariphagus indiens (0.0–8.3%) and Ch. infuscatelliis (0.0–1.1%). The parasitoid was collected in Ch. partellus and Ch. sacchariphagus indiens in all the four study years, whereas it was not observed in Ch. infuscatelliis during 1993–94. Parasitism rates in Ch. partellus were positively correlated with minimum temperature and afternoon RH. In Ch. sacchariphagus indiens, parasitism levels showed significant negative correlation with minimum temperature. Parasitism rates in Ch. saccharipliagus indidus larvae collected during harvest from seven sugar factory areas of Tamil Nadu State were lower in 1992–93 (0.0–1.8%) and higher in 1993–94 (0.0–3.2%) than those recorded at Coimbatore in the corresponding period (December-March). Cotesia flavipes was the only larval parasitoid recorded from the borers both at Coimbatore and the seven sugar factory areas surveyed in Tamil Nadu.

Two types of failures of attack by the top borer, Scirpophaga excerptalis Wlk., and their relation to sugarcane resistance to deadheart formation were studied in 164 genotypes of sugarcane for three years at two locations, viz., Karnal (Haryana State) and Chakia (Bihar State) in India. Attacks were unsuccessful either because of failure of neonate larvae in the mid-rib to reach the spindle (Type 1) or failure of older larvae in the spindle to reach the meristem (Type 2). Type 1 failure was frequent (19.2 to 26.7 %) while Type 2 was rare (2.9 to 5.1 %) among the genotypes. Larval dispersal and mortality were influenced by several factors related or unrelated to resistance and occurred concurrently in the same genotypes. This led to variation in larval mortality even among genotypes with the same levels of Type 1 failure. Those genotypes (Co 1223, Co 1224, E 276, Co 7224, Co 745, E 128, Co 62235, E 132, E 50, CoS 767 and CoJ 67) that had high scores of both Type 1 failure (14 to 18) and larval mortality (13 to 17) had the lowest percentages of deadhearts (11.8 to 24.4%).

Type 2 failure was not related to resistance; the highly resistant genotypes, Co 1223, Co 1224, E 276, Co 62235, E 132, CoS 767 and F133, showed no Type 2 failure. Furthermore, genotypes IA 3017 and Co 7322 which had the highest Type 2 failure scores of 9 and 12 respectively, had 61 and 36% deadhearts respectively. It is concluded that Type 2 failure may be caused by larval mortality in the spindles due to diseases and/or asphyxiation by accumulated rain water.

Studies were conducted on the spatial distribution of the sugarcane shoot borer, Chilo infuscatellus Snellen. Distribution was aggregated and described by the negative binomial model with a fixed mean size. Aggregated distribution probably resulted from environmental heterogeneity. The five quadrants of the field, i.e. north, south, east, west and central did not differ significantly in borer population. The optimum sample size, based on Iwao's formula, was 126 plants of 6 tillers each. The relevance of these findings to the management of C. infuscatellus is discussed.

The progress of the South African Sugar Association Experiment Station's 15-year-old biological control programme against Eldana saccharina Walker is reviewed. Two approaches at controlling this borer, a ‘new association’ and a ‘modified’ classical biological approach are summarised. Constraints identified as limiting the successful establishment of 10 egg, 12 larval and 2 pupal ‘new association’ parasitoids include host incompatibility, climatic incompatibility, poor ecological and biological life history knowledge, incorrect parasitoid searching ability and hyperparasitic tendencies. The 3 egg and 7 larval ‘modified’ classical biocontrol parasitoids which could be reared in the laboratory had constraints such as parasitising ability, different host habitat, differing host behaviours in different habitats and initial host identification difficulties to overcome. Difficulties encountered in this biological control programme against an indigenous pest insect show that biological control is not easy. Knowledge of constraints identified during this programme will hopefully be of use to others embarking on similar projects.

Studies on damage by larvae of Eldana saccharina Walker, Chilo zacconius Bleszynski and Sesamia spp. to sugar-cane stands revealed a high degree of correlation between internode damage and stalk damage as well as between internode damage and sugar content of the canes. For every 1% increase in stalk damage there was a corresponding increase of 0.214% in the internode damage. The distribution of larvae inside the internodes was found to depend on the level of infestation. At low field infestation level (below 25 larvae per 100 canes) the distribution was random. As the level of infestation increases, the distribution becomes more aggregated. The distribution of damage in the sugar-cane field was found to depend on the overall stalk damage. The distribution tends to be more contagious when field level of stalk damage was less than 45%. The damage became evenly distributed when stalk damage rose above 60%. An estimation of average loss of sugar caused by stem-borers to the variety NCO 376 (at the field damage rate of 8.6, 10.6 and 3.1% damage for bottom, middle and top parts respectively of the cane) was U.S.S332.10 per ha of field.