Mouse follicles were labeled with [3H]uridine and then cultured in vitro for 3 days. When oocytes were disrupted, about 40% of the total radiolabeled RNA could be sedimented at 9,000g. Fractionation of this RNA on poly(U)-Sepharose revealed that about 30% and 60% of the total amount of radiolabeled poly(A)- and poly(A)+ RNA, respectively, were in the pellet fraction. Treatments that disrupt protein structure reduced the amount of 9,000g sedimentable RNA and affected to the same extent the distribution of Poly(A)- and poly(A)+ RNA in the pellet and supernatant fractions. CsCl centrifugation of formaldehyde-fixed pellets revealed that virtually all of the radiolabeled RNA had a density significantly lower than that of ribosomes. The sedimentable RNA appeared not to be polysomal, membrane bound or associated wih a cytoskeleton. Agarose gel electrophoresis after poly(U)-Sepharose fractionation of either the pellet or supernatant revealed the presence of 28S, 18S, 5S + 4S, and heterodisperse poly(A)+ RNA. The size of distribution of poly(A)+ RNA in the pellet and supernatant fractions was fairly similar. Pulse-chase experiments revealed that the stability of poly(A)- RNA in the pellet and supernatant fractions was the same within the experimental error and a similar situation was found for poly(A)+ RNA. RNA in pellet translated in vitro coded for discrete size classes of protein. Since the relative band intensities were similar for both total and pellet RNA translated in vitro there seemed to be no major partitioning of specific size classes of mRNA into the pellet fraction. These results are discussed in terms of a possible composition of the lattice structures that accumulate during mouse oocyte growth and have been postulated to be a storage form for ribosome (Burkholder et al., '71).