Don't know if anyone's mentioned this data before, but I had to comment on the opinions that Grammars somehow enabled greater social mobility that comprehensive education does.
Anyone really interested in this should read 'state schools since the 1950s: the good news' by Adrian Elliot. The book (albeit a polemic) counters much of the anecdotal stuff one reads in the press about the parlous state of our schools.
I should probably admit a bias too, before we go on - until this year I was a secondary teacher in a successful state school.
There's a bit of data coming up, but I don't think one can really understand this debate unless one is given a bit of perspective.
So, Elliot cites several studies, including Jackson and Marsdenn (1966) who found that social class B (lower professional) children were 4.5 times more likely to complete a full seven year grammer school course than children from those from skilled manual classes. These children were in turn than 3.5 times more likely to complete grammar education than unskilled, class E children.
When they looked at their data in more detail, some of those working class children in groups D and E who went to grammar school were 'sunken middle class' - their families had formerly been in the middle classes, had owned businesses or had close relatives in middle class occupations.
Zweig (1961) found that only 8% of semi and unskilled workers' children passed the 11+. Coates and Silburn (1970) found that in St Annes, Nottingham only 1.5% of the children obtained a place in a grammar. They didn't study how many completed their courses.
Right - there you go - Grammars were highly selective places, which served to maintain inequality in society. Discuss!