At this point, I wouldn't say they're the most fluid groups, but they're the ones for which we have a greater problem to analyze because they appear and disappear from the provincial threat assessments a little more often; that impacts significantly on the national threat assessment.
Often enough, as Michel was explaining before, we rely significantly on the municipal, provincial, and other law enforcement agencies to provide this information to the provincial bureaus, and it then comes back to us. Often there is a cutoff point for the number of groups they report to us, so it's often the localized groups, we suspect, that are the ones coming down.
But there is a significant amount of fluidity. In the category 1 groups from this year, I think there's a change of perhaps 45% from the previous year. There is a significant turnaround, not only in the lower groups but in the higher-level groups as well.