days. I saw two on the 13th August flying together, and apparently just arrived, as they came from the sea. On the 16th I saw another. On the 21st I got a young Ruff from a flock of five, and the 6ame day several Green Sandpipers were seen at flooded pools amongst the sand-hills. They were very difficult to get near, as they always rise on seeing one approach, fly more unsteadily than a Snipe, and generally go straight into the air, seldom settling again within sight. On the 23rd I again saw several Green Sandpipers and wounded one, but did not get it. It settled on the open hill-side, as I expected dead, but rose quite wild again and flew to the beach,—the only occasion I ever saw one on the beach,—when it again rose suddenly and went away out of sight. On August 25th Green Sandpipers were still about. On the 28th, when looking for one, I saw a Colias Edusa butterfly, which I tried to catch, but failed to do so. No Green Sandpipers were seen after this. The first young Knots were killed on August 22nd, and a mature drake Pochard on the 30th. Mature Sanderlings appeared on the coast on the 16th August. I shot two (male and female) from different flocks, one of four, the other of nine birds. The old birds appear to pass on directly, for none are seen after the young arrive. Those obtained on their return from breeding are in a rather peculiar state, evidently changing from summer to winter plumage, but very little in the moult, the bare places on the breast being nearly obliterated, showing that some time has elapsed since they were sitting. These birds must have partially moulted on the back during the summer, the renewed feathers on the back having evidently come partially in summer plumage (that is faintly mottled), but as the feathers had grown they appear to have become grayer by degrees; and no doubt, from the broad edge of the new feathers and their indistinct markings, the remainder of the dark and cream-colour on each feather would wear out altogether as the season advanced and form part of the winter plumage of the bird. There is something curious in this species in this respect. I have met with old Sanderlings sometimes several weeks earlier. The first young bird of this species was obtained on August 18th. On the 14th September I got a Richardson's Arctic Gull whilst feeding on a dead herring on the shore. I did not before know they fed in this manner. The first young Bartailed Godwit was shot on the 19th September. Plenty were seen long before this, but in large flocks and wild. On August 16th I found red feathers of this bird cast where a flock had been preening themselves. An old male Bartailed Godwit, nearly in its plain gray winter plumage, but with numerous red feathers on its under parts and a few summer feathers remaining on its back, was killed on September 14th. It was moulting very regularly, the fifth primary from the end on each side being short, and the other four longest feathers were unchanged. This is the first bird I have seen killed here in a similar state of plumage, which is curious. The species lays four eggs, and therefore, if there were the
