The usefulness of serial study of C reactive protein in the early detection of neonatal septicaemia was evaluated in a neonatal unit using a commercially available latex agglutination slide test as a rapid screening method and electroimmunoassay as a reference method for C reactive protein determination. A positive latex test was obtained in 11 infants with verified septicaemia (positive blood culture), two infants with clinically evident infection but without bacteriological confirmation, one infant with recurrent chest infection due to Pseudomonas aeruginosa, and one infant who showed signs of birth asphyxia with meconium aspiration, but was not infected. Positive latex test results correlated with raised concentrations of C reactive protein, measured by immunoassay. In some instances, however, concentrations of C reactive protein in excess of 12 mg/100 ml gave weaker agglutination results in the slide test, which could be interpreted as negative results. In a sequential study of the infected infants, 6.3% of the values recorded on a slide test were false negatives. In contrast, false positive values were observed on a slide test in 1.9% of 27 non-infected infants. The higher percentage of false negative values may be due to the presence of excess antigen in the sera of some infected children. It is suggested that the latex test should be carried out on suitable dilutions of serum. Although the slide test may reliably indicate infection at an early stage in neonates, the C reactive protein response is non-specific, as seen in a non-infected infant who showed signs of birth asphyxia with meconium aspiration. Provided the non-specific nature of the C reactive protein response is recognised, the latex test may be a useful serum measurement for early diagnosis of neonatal septicaemia of the newborn. The test has the advantage of being performed easily, quickly, and cheaply.