Abstract The scope for the application of statistics to the field of rubberlike rnaterials is extensive, and various authors have already used statistical methods. Statistical methods derive the maximum amount of information from the results of an investigation, and at the same time supply a means of assessing the reliability and value of the information extracted. The effect of a variety of different blacks at various volume loadings on the properties of natural rubber and Neoprene-GN have been described. In the present paper the data are examined from other points of view, viz., the reproducibility of the results and the relations between the various properties. The main purpose of these investigations was to obtain the technological information already reported. At the same time, however, the program was designed as far as practicable to give information on the testing errors. Further useful information on the relation between the properties could be obtained by the application of the methods of statistical analysis. The program of experimental work with natural rubber covered a period of three months (November 1940–February 1941) and, to obviate as far as possible variations in the raw material, a quantity sufficient for the whole program was master-batched at the outset. The carbon blacks used were from normal supply, but special quantities were kept for this program. In 1942 a similar program was carried out with Neoprene-GN. The details of the mixes used, mixing procedure, and conditions of curing, have been given previously. Each mix was repeated three times on different days, and repeat determinations on the physical properties were carried out.