Spatial distribution of low-surface-brightness galaxies

Using redshift samples, we calculate the cross-correlation functions of LSBgalaxies with normal galaxies in complete samples (i.e. CfA and IRAS), whichenables us to compare directly the amplitudes and shapes of the correlationfunctions. For pair separations $r\gs 2\mpc$, we find that the shape is inagreement with that of the correlation functions for other galaxies. Theamplitudes ($A$) of $\xiab (r)$ are lower than those of the autocorrelationfunctions for the CfA and IRAS samples, with $A_{\rm LSB-CfA}:A_{\rmCfA-CfA}\approx 0.4$ and $A_{\rm LSB-IRAS}:A_{\rm IRAS-IRAS}\approx 0.6$. Theseresults suggest that LSB galaxies are imbedded in the same large scalestructure as other galaxies, but are less strongly clustered. This offers thehope that LSB galaxies may be unbiased tracers of the mass density on largescales. For $r\ls 2\mpc$, the cross-correlation functions are significantlylower than that expected from the extrapolation of $\xiab$ on larger scales,showing that the formation and survival of LSB galaxies may be inhibited byinteraction with neighboring galaxies. The model which matches these observations suggests that strong luminositysegregation in galaxy clustering is not a necessary consequence of biasedgalaxy formation, unless the effect of surface brightness is taken in toaccount. It is also implies a significant mass density in LSB galaxies.Comment: MNRAS in press, 20 pages Plain TeX, figures available upon request, abstract edited to avoid truncatio