From investigations previously reported it was evident that the failure of kernels to develop properly was at least in part a nutritional problem in which calcium played a vital role. Chemical analyses of peanut shells were made in an effort to obtain information on the mechanism whereby nutrient supply influences kernel development. Clean shells of 2-cavity size fruit were obtained from plants grown in five field experiments wherein widely different responses to various treatments had been obtained. The shells were divided into three groups, viz., those containing two, one, and zero developed kernels. Each group was analyzed for total nitrogen, potassium, magnesium, and calcium. Shells from four varieties were analyzed at two locations. Even though marked differences in the effect of treatment on yield and fruit quality had been obtained, there was no effect of treatment on the nitrogen, potassium, and magnesium contents of shells in any given kernel group. The 2-kernel shells, however, were considerably lower in these three constituents than were the 0-kernel shells, while the levels in the 1-kernel group were intermediate. This was found to be true regardless of treatment and was very consistent from location to location. In the two experiments in which the analyses of the shells of Virginia Bunch, N. C. Runner, Spanish 2B, and White Spanish were compared, the same general effect of kernel filling on composition was noted for all varieties and there was no marked differences in the composition of shells of the different varieties. In contrast to the results obtained with nitrogen, potassium, and magnesium, the calcium contents of the shells in a given group were markedly increased by the addition of calcium to the soil. There was no definite relationship between the calcium content of shells and kernel development. Analyses of a bulk sample of clean 2-cavity shells (all kernel groups combined) at one location showed a marked effect of calcium treatment. The nitrogen, potassium, and magnesium contents were decreased and the calcium content was increased by the additions of gypsum. The beneficial effect of calcium in producing a high proportion of 2-kernel fruit was thought to be responsible for these differences. Kernel analyses were made at one of the experiments and the contents of all constituents except calcium was found to be higher in the kernel than in the shell. It was postulated therefore that the decrease in the nutrient content of shells of the 2-kernel groups was due to the movement of the nutrients from the shells into the developing kernels.