Anderson et al. have found that experimental values for gravitational constant, G, are oscillatory with a period of P = 5.899 +- 0.062 yr and an amplitude of A = (0.001619 +- 0.000103) F, where F is a factor that is the inverse of the 11 powers of ten expressed in SI Base Units. The period P correlates with the 5.9 year periodic signal in length of day (LOD). Previously it was assumed that changes in LOD follow from fluid core motions and inner-core coupling. But the origin of the G/LOD correlation is still not explained. Next, Schlamminger et al. have found that corrections to the G data reported by Anderson et al. significantly weaken the G/LOD correlation. But in Table III there appears the fit sinusoidal function for G with period fixed at 5.93 year. Here, applying the Scale-Symmetric Theory (SST), we show the origin of the periodic changes in LOD and the illusory changes in G and the origin of the experimental amplitude of G that is illusory as well. The calculated period of LOD and illusory period of G are both 5.941 yr whereas illusory amplitude of G is 0.00183 F. Here, the 5.941 year periodic signal in LOD we interpret as due to advection of the virtual cloud, composed of the virtual electron-positron pairs produced by Earth, caused by its interactions with the dark-matter (DM) structures that overlap with the Earth orbit. Such advection follows from the confinement and entanglement of virtual electrons and positrons and DM structures. The advection causes periodic increase in mass of Earth and measuring apparatus. The amplitude of the changes in mass of Earth follows directly from masses of nucleons (we must take into account the nuclear binding energy) and electrons. The changes in mass lead to the illusion of periodic changes in G. The SST results are consistent with experimental data. Moreover, SST shows that the real changes in G can be infinitesimal only.