(1) The variations due to change in arc length, in the relative intensities of lines in the ultra-violet spectra of different direct current arcs, have been studied with the aid of the quartz spectrograph. (2) Spectrograms of the solid carbon arc for several electrode distances at a power input of 400 watts were examined in detail with a microphotometer. The measurements showed that the intensity varied with arc length in a different manner for each line or group of lines examined. (3) Short arcs were found to be deficient in radiation of short wave-length but were strong in the visible and near ultra-violet. Long arcs were richer in ultra-violet radiation of short wave-length. (4) A comparison of the spectrograms of various arcs with the results obtained with the quartz-aluminium photo-electric cell for the same arcs showed that the cell followed closely the intensity variations of the radiations in the far ultra-violet. (5) All arcs investigated—carbon, iron, graphite, tungsten, and aluminium-cored—showed (a) that the relative intensities of the spectral lines varied with arc length when the power input was kept constant, and (b) that the extension of the spectra at the short wave end was a maximum at an arc length which had approximately the value L (Imax) found by the aluminium cell measurement. (6) The aluminium cell is extremely sensitive to radiation from the aluminium-cored carbon arc. The arc length giving maximum intensity of emission in the far ultra-violet is critical, the intensity falling off rapidly on both sides of the maximum. (7) The form of the arc length-intensity curve depends upon the particular method of measurement employed and the portion of the spectrum to which the measuring instrument responds. (8) It is emphasised that it is impossible to generalise on the basis of results obtained with any one selective measuring device. The work of Griffith and his collaborators is criticised on this ground. (9) The mechanism of the arc discharge is briefly considered. We desire again to record our indebtedness and to express our sincere thanks to Mr. Thorne Baker, of the Imperial Dry Plate Co. Ltd., for the loan of the Hilger quartz spectrograph which enabled us to carry out the investigations described in this and previous papers on the electric arc.