Sixthly, the saving of the time usually lost in preparation for swarming, when the bees hang inactively in clusters, on the outsides of the hives, for many days, sometimes for weeks, particularly if the weather be unfavourable. It seems right to remark in this place, that though this _clustering_ or _hanging out_ of the bees is generally regarded as one of the strongest symptoms of an approaching swarm, it is nevertheless a deceptive one. It does certainly indicate that there are bees sufficient to throw off a swarm, and is sometimes evidence of an anxiety to do so; but unless there be a queen ready to go off with them, however distrest for room, the clustering will sometimes continue for a considerable time; in hot dry seasons it may last till the middle of August. This clustering, as before observed, is very prejudicial, as it causes the bees to be inactive in their principal harvest season, when every bee ought to be fully employed, and may induce a habit of inactivity for the future. Clustering likewise obstructs the operations of the bees that are active, by interrupting the thoroughfare to the hive. These disadvantages are admirably remedied by storifying, without which, independently of the loss of time to the bees, a constant system of watchfulness must be kept up by the proprietor, during the whole period of the bees clustering out, otherwise a swarm may be lost. Storifying, though generally, is not invariably successful in causing the clustered bees to reenter the boxes: where it fails to do so, if a young queen were ready to assume the sovereignty of the colony, the clustered bees would swarm and seek a new habitation with the old one. M. Reaumur drowned several hives thus circumstanced, and examined all their inmates most minutely, but could never find more than a single queen, and this the old one; in none of these hives did he find royal larvæ.