The continuous nucleation of bcc solid 3 He from the liquid phase due to the presence of a MgO substrate is detected. For superfluid 4 He there is no evidence of continuous nucleation of either the hep or the bcc solid from the substrate. The experiment contrasts the influence of the square lattice of MgO with the triangular lattice of a basal-plane graphite substrate. The phenomenon of pore condensation for solid 3 He is discussed.PACS numbers: 67.80. Cx, 68.45.Da In this paper we present new results which compare the onset of solidification of 3 He and 4 He on a magnesium oxide substrate, which has cubic symmetry, with that already reported 1 * 2 for solidification on Grafoil, which has a hexagonal symmetry. The Grafoil experiments had shown that hep 4 He undergoes continuous solid nucleation whereas, surprisingly, bcc 3 He did not nucleate similarly. Since the van der Waals forces between the substrate and either isotope are identical, and this force is considered 1 to be responsible for the superpressurization of the first few helium layers on the substrate, we were led to suggest that the substrate symmetry (hexagonal graphite encouraging the hep 4 He) is the determining factor in the onset of solidification. To test this conclusion we repeated the experiments with a MgO substrate 3 * 4 which has a cubic symmetry. For this substrate we observed uniform nucjeation of the bcc 3 He whereas 4 He does not show any signs of surface solidification in either hep or bcc regions.We should emphasize that the experiment on Grafoil alone did not prove the relevance of substrate symmetry. During the condensation of the first layer of He on Grafoil, it is known from adsorption studies 5 that initially an ordered state is reached in which the helium atoms are registered with the graphite lattice. Subsequently, as the coverage is increased, a close-packed triangular two-dimensional lattice is formed, with a spacing incommensurate with the graphite lattice. It can then be argued that this layer, which is formed by either isotope, is the natural base for the continuing growth of hep 4 He, whereas it would not encourage the further growth of a bcc solid. The present experiments disprove the universality of this model, and demonstrate the importance of a matching substrate symmetry. Further experimental evidence from direct observation of the growth of aligned 4 He crystals on the Grafoil substrate will also be presented to support the conclusion.The experimental evidence for the onset of solidification is deduced from the measurement of equilibrium points along isopycnals 6 in the vicinity of the bulk melting curve. There appears a substantial difference between the isopycnal and the isochore for the bulk liquid when surface solidification occurs, in the presence of a substrate of sufficiently large surface area. The present experiments used a 2.9-g MgO sample prepared from compressed smoke at the University of Washington. 3 Its properties have been measured by Dash etal. 4 who showed that it produces krypton step like adsor...