A couple of answers to this. Both in the form of links:
First to
It asks the question whether there is an objective way to count crowds. It concludes:
'In fact, there is. Clark McPhail, professor emeritus of sociology at the University of Illinois at Urbana-Champaign, is one of America's preeminent authorities on protest crowds. His award-winning 1991 book on crowd behavior, "The Myth of the Madding Crowd," is required reading in his field.'
And second to
--
Date: Tue, 18 Feb 2003 13:41:01
A further (and slightly lateral) measure is provided in today's
Guardian:
"This month's Guardian/ICM poll also shows that at least one person from
1.25 million households in Britain went on Saturday's anti-war march in
London, confirming estimates that between one million and two million
people went on the march."
from http://www.guardian.co.uk/Iraq/Story/0,2763,897852,00.html
Although these polls are based on a telephone sample of 1,000 people...