2011MNRAS.414..445S

Recent reports suggest that elliptical galaxies have increased their size dramatically over the last ∼8 Gyr. This result points to a major rethink of the processes dominating the late-time evolution of galaxies. In this paper we present the first estimates for the scale sizes of brightest cluster galaxies (BCGs) in the redshift range 0.8 < z < 1.3 from an analysis of deep Hubble Space Telescope imaging, comparing to a well-matched local sample taken from the Local Cluster Substructure Survey at z∼ 0.2. For a small sample of five high-redshift BCGs we measure half-light radii ranging from 14 to 53 kpc using de Vaucuoleurs profile fits, with an average determined from stacking of 32.1±2.5 kpc compared to a value 43.2±1.0 kpc for the low-redshift comparison sample. This implies that the scale sizes of BCGs at z= 1 are ≃ 30 per cent smaller than at z= 0.25. Analyses comparing either Sérsic or Petrosian radii also indicate little or no evolution between the two samples. The detection of only modest evolution at most out to z= 1 argues against BCGs having undergone the large increase in size reported for massive galaxies since z= 2 and in fact the scale-size evolution of BCGs appears closer to that reported for radio galaxies over a similar epoch. We conclude that this lack of size evolution, particularly when coupled with recent results on the lack of BCG stellar mass evolution, demonstrates that major merging is not an important process in the late-time evolution of these systems. The homogeneity and maturity of BCGs at z= 1 continues to challenge galaxy evolution models.