An understanding of the radiation emitted by impurities within a plasma is crucial if spectral line intensities are to be used in detailed studies, such as the analysis of impurity transport. The simplest and most direct check that can be made on any measurements of line intensities is to analyse their ratios, this avoiding uncertainties as to the plasma volume emitting the radiation, the absolute sensitivity calibration of the spectrometer and, in some cases, the need for accurate measurements of parameters such as electron density. A comparison of measured and modelled line intensity ratios can also lead to instrumental sensitivity calibrations and this has been done for the JET VUV SPRED survey spectrometer by Lawson et al. (2008). At short wavelengths, 100Å­390Å, this was achieved by comparing the measured and modelled spectral line intensity ratios of a number of Na- and Li-like doublets. The agreement between the measurements and the modelled ratios was good, being of the order of the measurement accuracy, ~6%; this resulted in a particularly accurate sensitivity calibration. At longer wavelengths, 390Å­1000Å, a similar approach using C line intensity ratios was tried, C being the main low Z intrinsic impurity in JET. However, the agreement between the measured and modelled C ratios was found to be poorer, with discrepancies of up to ~45%. So as to repeat the accuracy of the short wavelength calibration at longer wavelengths and to gain understanding of the interpretation of spectral line intensity measurements from the plasma edge, a detailed study has been made of the C line intensity ratios. The most thorough analysis has been carried out for the CIV ionization stage, the highest temperature and, therefore, innermost ion always found in the Scrape-Off Layer (SOL) of the JET plasma. Comparisons for CIII and CII are also discussed. The observations have been made using a SPRED and a Schwob-Fraenkel spectrometer, which have near-horizontal lines of sight viewing the core and SOL plasmas. In the present study, ohmic, limiter pulses have been considered. The discrepancies between the measured and modelled C line intensity ratios are described and various explanations have been investigated. Line blending is the commonest reason for disagreements in the VUV and a full application of the collisional-radiative model requires the inclusion of recombination, particularly from charge exchange collisions. The steady state assumption normally used in the collisional-radiative model has been checked and the possibility of the electron energy distribution being non-Maxwellian is being considered. Of these, only the non-Maxwellian distributions allow better agreement to be achieved, although it is not clear at this stage whether the improvement is fortuitous or a valid explanation of the discrepancy. The effect of line-of-sight integration through plasma regions with steep gradients has been addressed by Zacks (2008) and may provide an explanation of the discrepancy.