In order to investigate the suggestion of von Montigny et al. [1995b] that
the jets in EGRET-identified
radio sources may be preferentially straight, information on the mas-scale and arcsecond-scale
radio emission from EGRET-identified sources has been compiled from the literature. 18 of the 40
high-confidence EGRET identifications listed by Thompson
et al. [1995] were found to have the relevant information of
sufficient quality to find the difference between the
mas-scale and arcsecond-scale radio structure position
angles. For the remaining 22 sources either no information was found,
or else the information was not of a sufficient quality to allow the
determination of the position angle difference. The data compiled for the 18 sources are plotted as a histogram in Figure 4.13. From the histogram it can be seen that a strong peak in the
distribution of position angle differences occurs near 0 
, or
no misalignment.
Figure 4.13: Misalignment angles for 18 EGRET-identified radio sources
To investigate the possibility that the core dominated radio sources not identified by EGRET preferentially have bent jets, information on the mas-scale and arcsecond-scale radio structure of 11 core dominated quasars and BL Lac objects from the list given by Vermeulen and Cohen [1994] and 4 core dominated quasars appearing in both Murphy, Browne, and Perley [1993] and Polatidis et al. [1995] was collected. In addition, 3 of the sources considered for the first time here, PKS 0438-438, 0637-752, and 1514-241 were utilised to complete a list of 18 objects. Figure 4.14 is the histogram of the position angle differences for these 18 sources.
Figure 4.14: Misalignment angles for 18 radio sources not identified by EGRET
A simple Kolmogorov-Smirnov two-sample test was used to determine if the differences between the distributions of position angle differences are significant. For 18 objects and a maximum difference of 4/18 in the cumulative distribution it is not possible to reject the null hypothesis that the two samples are drawn from the same parent distribution. On the basis of these 36 objects there is no reason to conclude that EGRET-identified radio sources have a greater preference for straight jets than radio sources not identified by EGRET.