We have begun to examine the EGRET database for short term variations in the fluxes of blazars and unidentified sources at high Galactic latitude. We find that several AGN show previously unreported variability. Such variations are consistent with inverse Compton scattering processes in a shock propagating through a relativistic jet.

The largest group of sources identified by EGRET are Blazars. This sub-class of AGN is well known to vary in flux in all energy bands on time-scales ranging from a few minutes (in the optical and X-ray bands) up to decades (radio and optical regimes). In addition to variations of the gamma-ray flux between different viewing periods, the brightest of these sources showed a few remarkable gamma-ray flares on time-scales of about one day, confirming the extension of the ``Intraday-Variability (IDV)'' phenomenon into the GeV range. We present first results of a systematic approach to study fast variability with EGRET data. This statistical approach confirms the existence of IDV even during epochs when no strong flares are detected. This provides additional constraints on the site of the gamma-ray emission and allows cross-correlation analyses with light curves obtained at other frequencies even during periods of low flux. We also find that some stronger sources have fluxes systematically above threshold even during quiescent states. Despite the low count rates this allows explicit comparisons of flare amplitudes with other energy bands.

We present the results of the multi-wavelength campaigns on 3C 273 in 1993-1995. During the observations in late 1993 this quasar showed an increase of its flux for energies greater equal 100 MeV from about 2.1 x 10^-7 photons/cm^2/s to approximately 5.6 x 10^-7 photons/cm^2/s during a radio outburst at 14.5, 22 and 37 GHz. However, no one-to-one correlation of the gamma-ray radiation with any frequency could be found. The photon spectral index of the high energy spectrum changed from (3.20 +- 0.54) to (2.20 +- 0.22) in the sense that the spectrum flattened when the gamma-ray flux increased. Fits of the three most prominent models (synchrotron self-Comptonization, external inverse Comptonization and the proton initiated cascade model) for the explanation of the high gamma-ray emission of active galactic nuclei were performed to the multi-wavelength spectrum of 3C 273. All three models are able to represent the basic features of the multi-wavelength spectrum. Although there are some differences the data are still not decisive enough to discriminate between the models.

The possibility of a correlation between the radio (cm)- and $\gamma$-ray luminosity of variable AGN seen by EGRET is investigated. We performed Monte-Carlo simulations of typical data sets and applied different correlation techniques (partial correlation analysis, $\chi^2$-test applied on flux-flux relations) in view of a truncation bias caused by sensitivity limits of the surveys. For K-corrected flux densities, we find that with the least squares method only a linear correlation can be recovered. Partial correlation analysis on the other side provides a robust tool to detect correlations even in flux-limited samples if intrinsic scatter does not exceed $\sim 40$ \% of the original $\gamma$-ray luminosity. The analysis presented in this paper takes into account redshift bias and truncation effects simultaneously which was never considered in earlier papers. Applying this analysis to simultaneously observed radio- and $\gamma$-ray data, no correlation is found. However, an artificial correlation appears when using the mean flux. This is probably due to the reduction of the dynamical range in the flux-flux relation. Furthermore, we show that comparing the emission in both spectral bands at a high activity state leads to no convincing correlation. In conclusion, we can not confirm a correlation between radio and $\gamma$-ray luminosities of AGN which is claimed in previous works.