The dredge descended with the extirpators pointing upward. When their heads reached the level of the upper floor, the tools were removed. Then the dredge was raised again. In this way the extirpators lay upon the floor, and, if the lifting was continued, they placed themselves in their working position, in which they were fixed by the bolts A" B" C" (Fig. 1). After this, the apparatus was let down and revolved. The rakes divided the earth, the scrapers collected it, and the bags pocketed it. The great difficulty was to cause the tubbing to descend vertically, and also to overcome the enormous lateral pressure exerted upon it by the earth that was being traversed. Water put into the shaft helped somewhat, but the great stress to be exerted had to be effected by means of powerful jack screws. These were placed directly upon the tubbing, and bore against strong beams whose extremities were inserted into the masonry. As a usual thing it is not easy to use more than four or six such jacks, since the number of beams that can be employed is limited, owing to the danger of obstructing the mouth of the shaft. Yet twelve were used by Mr. Chavatte, and this number might have been doubled had it been necessary. As we have seen, the frame, K K (Pl. 1, Fig. 3), was provided with an oak circle traversed by 32 bolts. The length of these latter was two inches and a quarter longer than they needed to have been, or they were provided with wooden collars of that thickness. Later on, these collars were replaced with iron bars that held the wood against which the jacks bore in order to press the tubbing downward (Pl. 1, Figs. 10, 11, 12, and 13). Mr. Chavatte's great anxiety was to know whether he should succeed in causing the first section of tubbing to traverse the four feet of gravel; for in case it did not pass, he would be obliged to employ a second section of smaller diameter, thus increasing the expense. He was persuaded that the coarse gravel remaining in the side of the shaft would greatly retard the descent of the tubbing. So he had decided to remove such obstructions at the proper moment through divers or a diving bell. Then an idea occurred to him that dispensed with all that trouble, and allowed him to continue with the first section. This was to place upon the dredge two claw-bars, T (Pl. 2, Fig. 3), which effected the operation of widening with wonderful ease. To do this it was only necessary to turn up the bags, and revolve the apparatus during its descent. The claw at the extremity of the bar pulled out everything within its reach, and thus made an absolutely free passage for the tubbing. The sands and gravels were passed by means of a single section of tubbing 31 feet in length, which was not stopped until it had penetrated a stratum of white chalk to a depth of two yards. This chalk had no consistency, although it contained thin plates of quite large dimensions. These were cut, as if with a punch, by means of the teeth of the extirpator.