The salt 1,1,3,3-tetramethylguanidinium bromide, [((CH3)2N)2C=NH2]+Br- or [tmgH]Br, was found to melt at 135(5) °C, forming what may be referred to as a moderate temperature ionic liquid. The chemistry was studied and compared with the corresponding chloride compound. We present X-ray diffraction and Raman evidence to show that also the bromide salt contains dimeric ion pair “molecules” in the crystalline state and probably also in the liquid state. The structure of [tmgH]Br determined at 120(2) K was found to be monoclinic, space group P21/n, with a = 7.2072(14), b = 13.335(3), c = 9.378(2) Å, beta = 104.31(3)°, Z = 2, based on 11769 reflections, measured from θ = 2.71-28.00° on a small colorless needle crystal. Raman and IR spectra are presented and assigned. When heated, both the chloride and the bromide salts form vapor phases. The Raman spectra of the vapors are surprisingly alike, showing, for example, a characteristic strong band at 2229 cm-1. This band was interpreted by some of us to show that the [tmgH]Cl gas phase should consist of monomeric ion pair “molecules” held together by a single N-H+ · · · Cl- hydrogen bond, the stretching vibration of which should be causing the band, based on ab initio molecular orbital density functional theory type calculations. It is not likely that both the bromide and chloride should have identical spectra. As explanation, the formation of 1,1-dimethylcyanamide gas is proposed, by decomposition of [tmgH]X leaving dimethylammonium halogenide (X = Cl, Br). The Raman spectra of all gas phases were quite identical and fitted the calculated spectrum of dimethylcyanamide. It is concluded that monomeric ion pair “molecules” held together by single N-H+ · · ·X- hydrogen bonds probably do not exist in the vapor phase over the solids at about 200-230 °C.