We re-examine Voyager Ultraviolet Spectrograph data used to establish upper limits to the \lambda 500–900 Å and \lambda 900–1100 Å cosmic diffuse background .
The measurement of diffuse flux with the Voyager UVS data requires complex corrections for noise sources which are far larger than the astronomical signal .
In the analyses carried out to date , the upper limits obtained on the diffuse background show statistical anomalies which indicate that substantial systematic errors are present .
We detail these anomalies and identify specific problems with the analysis .
We derive statistically robust 2 \sigma upper limits for continuum flux of 570 photons s ^ { -1 } cm ^ { -2 } ster ^ { -1 } Å ^ { -1 } and for the 1000 Å diffuse line flux of 11,790 photons s ^ { -1 } cm ^ { -2 } ster ^ { -1 } .
The true limits may be substantially higher because of unknown systematic uncertainties .
The new statistical limits alone are insufficient to support previous conclusions based on the Voyager data including work on the character of interstellar dust and estimates of the diffuse extragalactic far UV background as absorbed by intergalactic dust .