Under given experimental conditions, a reproducible sequence of biochemical events can be measured during sporulation in Bacillus thuringiensis. Electron microscope studies reveal that, during its formation, the crystalline inclusion of Bacillus thuringiensis is closely associated with the exosporium. This association persists through several serial sections and can also be demonstrated in a mutant strain which is biochemically blocked soon after crystal formation.Two asporogenic crystalliferous strains produce a toxin active on larvae of Plutella maculipennis whereas two sporogenic, acrystalliferous strains are not toxic.Data from immunological assays and from pulse-chase and label-chase experiments with [14C]leucine indicate that the crystal protein is synthesized at the time of appearance of crystal antigens. It is suggested that the crystal is synthesized and assembled on the exosporium.The production by Bacillus thuringiensis of a crystalline parasporal body concomitant with the formation of the endospore is the principal distinction between strains of B. thuringiensis and of Bacillus cereus. Much work on B. thuringiensis has centred on its entomodical capacity which has been attributed, in part, to the parasporal inclusion, although the molecular nature of the toxic agent and its mode of action are as yet unknown. Several workers have studied the formation of the crystal and its relationship to the spore. Young and Pitz-James [I] described the simultaneous development of the spore and crystal as seen in the light microscope and followed changes in a number of biochemical parameters during sporulation. Crystal antigens, which are absent from vegetative cells [2,3], appear during sporulation [2,4] some time before the development of heat resistance. Monro [5] provided evidence from radioactive labelling studies that the bulk, if not all, of the crystal protein is formed de novo during sporulation and concluded from immunological comparison [2] of crystal protein with soluble spore protein that the crystal and the spore did not share common components. Recent studies [3,6] have shown, however, that the crystal protein is immunologically and biochemically similar to a substantial protein fraction which can be removed from spores by extraction with reagents which dissolve the crystal protein. The precise extent and nature of this similarity remain to be established. Treatment of whole spores with urea-mercaptoethanol [7] removes some of this protein and produces spores which have altered germination requirements and different sensitivity to lysozyme, while retaining viability and heat resistance.It thus appears that the crystal protein may have some physiological function in the spore and that crystal formation may result from uncontrolled synthesis of spore protein.I n Bacillus subtilis it has been established that, under given conditions, there is a definite sequence of biochemical [8-101 events during sporulation. I n several spore-forming bacteria, including B. cereus [Ill and Bacillus subtilis [12,13], t...