We use the two-point correlation function to calculate the clustering properties of the recently completed SSRS2 survey , which probes two well separated regions of the sky , allowing one to evaluate the sensitivity of sample-to-sample variations .
Taking advantage of the large number of galaxies in the combined sample , we also investigate the dependence of clustering on the internal properties of galaxies .

The redshift space correlation function for the combined magnitude-limited sample of the SSRS2 is given by \xi ( s ) = ( s / 5.85 h ^ { -1 } Mpc ) ^ { -1.60 } for separations between 2 \leq s \leq 11 h ^ { -1 } Mpc , while our best estimate for the real space correlation function is \xi ( r ) = ( r / 5.36 h ^ { -1 } Mpc ) ^ { -1.86 } .
Both are comparable to previous measurements using surveys of optical galaxies over much larger and independent volumes .
By comparing the correlation function calculated in redshift and real space we find that the redshift distortion on intermediate scales is small .
This result implies that the observed redshift-space distribution of galaxies is close to that in real space , and that \beta = \Omega ^ { 0.6 } / b < 1 , where \Omega is the cosmological density parameter and b is the linear biasing factor for optical galaxies .

We have used the SSRS2 sample to study the dependence of \xi on the internal properties of galaxies such as luminosity , morphology and color .
We confirm earlier results that luminous galaxies ( L > L ^ { * } ) are more clustered than sub- L ^ { * } galaxies and that the luminosity segregation is scale-independent .
We also find that early types are more clustered than late types .
However , in the absence of rich clusters , the relative bias between early and late types in real space , b _ { E + S 0 } / b _ { S } \sim 1.2 , is not as strong as previously estimated .
Furthermore , both morphologies present a luminosity-dependent bias , with the early types showing a slightly stronger dependence on the luminosity .
We also find that red galaxies are significantly more clustered than blue ones , with a mean relative bias of b _ { R } / b _ { B } \sim 1.4 , stronger than that seen for morphology .
Finally , by comparing our results with the measurements obtained from the infrared-selected galaxies we determine that the relative bias between optical and IRAS galaxies in real space is b _ { o } / b _ { I } \sim 1.4 .