Sometimes, when the current is light, the fisherman prepares a “mule” to assist the boat in towing the resisting drag. This “mule” consists of a wooden frame, hinged in V-shape, and is fastened several feet in advance of the boat with the V end pointed down the stream. It sinks low in the water, and the current pressing against the angle carries it along, and thus tows the skiff and the resisting drag at a uniform rate of speed. When there is not sufficient current even for this contrivance, as in the wide reaches and in the lakes, oars, sails, and even power engines may be used for propelling the boat. As the crowfoot drag is slowly drawn along the bottom, it comes in contact with the mollusks feeding with open shells. When a hook or other part of the drag enters an open shell, the mollusk immediately closes firmly upon the intruding object and clings thereto long enough to be drawn up into the boat. In this way, where the Unios are thick, nearly every hook becomes freighted, and some may have two or three shells clinging thereto. It is easy to collect fifty mollusks in passing over a length of two hundred feet. Two drags are carried by each fisherman, and the second one is put overboard as soon as the first one is ready to be raised. This is suspended with the bar across two upright forks on either side of the boat with the prongs swinging freely, and the mussels are removed therefrom. When this operation is completed, the drag is put overboard and the other one is ready for lifting. This apparatus is very effective, and as much as a ton of shells has been taken by one man in twelve hours, but the average is very much less, probably not over four or five hundred pounds. Objection is made to this manner of fishing, since many mollusks not brought to the surface are so injured that they die. A cruder implement of similar type has long been employed on many logging streams. The weighted branch of a tree is dragged on the bottom behind a raft of logs, and the mussels attach themselves to the twigs in the same manner as on the crowfoot hooks. During the pearling excitement in Arkansas, a considerable portion of the choice pearls were found, not in the mussels, but lying loosely in the mud of the shores, indicating that under some circumstances, as agitation by freshets or floods, the loose pearls are shaken out from the Unios. In some instances, indeed, the pearls were found upon or in the soil at some distance from streams or lakes. It is reported that in October, 1897, Mr. J. W. McIntosh, of the northern part of Lonoke County, while digging post-holes in the old bed of Cypress Bayou, found a number of pearls, some “as large as a 44–caliber Winchester ball,” lying within the shells at a depth of a foot and a half below the surface. This peculiar occurrence is partly explained by the wide extension of the waters in flood times over the low region, and by the shifting of streams and the isolation of cut-offs.