174 creek, using water from its extreme upper part, and the other near the mouth, using the creek water and water brought from Amador creek by a ditch. During the past few years only the upper mine has been worked and the operations, owing to insufficient water, have been limited to a few weeks. The results have not been so profitable as formerly, because the part of the rock channel, or bench, mined during the past few years has in places a considerable filling of boulder clay, which renders hydrau- licking difficult. Hydraulicking was carried on near the mouth in 1920 and 1921, but was not very successful because of the high banks on both sides and the difficulty of reaching bedrock, and obtaining sufficient grade for the sluice boxes. There is evidence that a rock channel, beginning at a point on Perkins creek 300 feet above Lightning Creek flat, extends west beneath a high drift hill. The possible occurrence of an old channel parallel to Lightning creek at this locality was long ago recognized by the miners and a great deal of work has been done in search of its continuation. The conditions at this locality are unusua! and well illustrate the difficulty of tracing the old rock channels that are in many places buried beneath thick drift deposits, the surface contours of which give little or no indication of the bedrock contours. A tunnel was run west in the channel or depres- sion in the bedrock and the bedrock was found to dip towards the west. Later, a tunnel 50 feet long, which had the rim-rock on the south side and the posts of the old Chinese tunnel on the north side, was run by Sparks and Felker. They obtained some fine gold, including a few pieces up to $1.50 in value, and found that there was a grade downstream which prevented drainage of the ground. A number of prospect pits and a tunnel driven from the bank of Lightning creek seemed to show that bedrock occurs near the surface along the north bank of Lightning creek down to the next small creek, known as the Goat Ranch, 1,000 feet below, on which two shafts were sunk in search of the channel. The lower one of these was sunk in the early days and appears to have been partly in bedrock. The other was sunk by Harry Jones and is 60 feet deep to bedrock. From the mouth of the Sparks and Felker tunnel to the bottom of the Jones shaft is a fall of 65 feet, the distance between the two points being 875 feet. Assuming that the tunnel and the Jones shaft are in the same rock channel, the average grade between the two points would be over 7 per cent, which seems rather high, so that it is doubtful whether the Jones shaft is located over the channel or where the channel joins Lightning creek. Apparently, little gold was found in the Jones shaft or in the older one nearby, as little drifting was done. The fact that the bedrock channel of Perkins creek in its lower part is not graded to the bottom of Lightning Creek channel, whereas the channels of Amador and Chisholm creeks are so graded, favours the view that a channel other than that of the present creek at its mouth exists and, where it joins Lightning creek, is graded to the bottom of the deep channel. Fountain Creek Fountain creek or Fontaine creek, as it is said to have been formerly known, flows southwest into Swift river, and in its upper part becomes two streams, one heading in a flat-bottomed pass near the head of Peters