W. A. Philipps, Two-level states in glasses, Reports on Progress in Physics, vol.50, issue.12, p.1657, 1987.
DOI : 10.1088/0034-4885/50/12/003

A. J. Lock, T. M. Creemers, and S. Völker, Spectral diffusion in glasses under high pressure: A study by time-resolved hole-burning, The Journal of Chemical Physics, vol.110, issue.15, p.7467, 1999.
DOI : 10.1063/1.478649

J. Joffrin and A. Levelut, Virtual phonon exchange in glasses, Journal de Physique, vol.36, issue.9, p.811, 1976.
DOI : 10.1051/jphys:01975003609081100
URL : https://hal.archives-ouvertes.fr/jpa-00208320

P. Neu, D. R. Reichman, and R. J. Silbey, Spectral diffusion on ultralong time scales in low-temperature glasses, Physical Review B, vol.56, issue.9, p.5250, 1997.
DOI : 10.1103/PhysRevB.56.5250

H. M. Carruzzo, E. R. Grannan, and C. C. Yu, Nonequilibrium dielectric behavior in glasses at low temperatures: Evidence for interacting defects, Physical Review B, vol.50, issue.10, p.6685, 1994.
DOI : 10.1103/PhysRevB.50.6685

C. Enss and S. Hunklinger, Incoherent Tunneling in Glasses at Very Low Temperatures, Physical Review Letters, vol.79, issue.15, p.2831, 1997.
DOI : 10.1103/PhysRevLett.79.2831

E. Thompson, Low Temperature Acoustic Properties of Amorphous Silica and the Tunneling Model, Physical Review Letters, vol.84, issue.20, p.4601, 2000.
DOI : 10.1103/PhysRevLett.84.4601

S. N. Coppersmith, Frustrated Interactions and Tunneling: Two-Level Systems in Glasses, Physical Review Letters, vol.67, issue.17, p.2315, 1991.
DOI : 10.1103/PhysRevLett.67.2315

A. L. Burin, Y. Kagan, I. Ya, and . Polishchuk, Energy Transport Induced by an External Alternating Field in Strongly Disordered Media, Physical Review Letters, vol.86, issue.24, p.5616, 2001.
DOI : 10.1103/PhysRevLett.86.5616

X. Experiments, carried out onto layers deposited on Si-111 substrates, give the two main features of amorphous solids: (i) the layers do not yield any Bragg peak; (ii) setting the angle of incidence ˆ 5 and scanning the detector angle, each layer gives a ''flattened hump'' ranging from 15 to 35 This flattened hump is mainly unchanged when h ranges from 40 nm to 1 m. Since the ''hump's height/noise'' ratio worsens at small h, we cannot exclude that this ''hump'' is somehow deformed at small h. Should this be the case, it might lead to a small variation of the elementary dipole p 0 (or of the TLS density of states P P) with h. This is of no consequence here since, the standard noninteracting TLS model