never seen an adult male at this season. Mr. Allan Brooks is of the opinion that the males go directly to the coast at this time.
The courtship display is witnessed in the flocks just prior to their splitting up into pairs. It is attended by much solemn bowing on the part of the drake, with a frequent backward kick, sufficiently strong to send a jet of water several feet into the air. His violet head is puffed out to the greatest possible extent, and altogether he is a handsome bird as, in a frenzy of sexual excitement, he swims up to the soberly attired duck. Sometimes the entire flock will commence to feed as if at a given signal, and again all the birds will simultaneously take wing and circle about the lake several times before once more splashing down to resume their courtship.
By May 1 all breeding birds are mated and scattered over the country, seldom more than one or two pairs on a lake. The Barrow Golden-eye shows a marked predilection for lakes that are strongly alkaline, even if they are poor in aquatic vegetation and in the midst of an open country with the nearest timber a half mile or more away. Such lakes are rich in small crustaceans, the chief food of this duck, and no doubt the lakes are occupied on account of the food provided, without reference to the availability of nesting sites.
An abandoned flicker's hole is usually selected for. the nest, frequently in a dead yellow pine, for in this tree decay is rapid, and the hole soon becomes much enlarged. One can generally tell if the hole is occupied, by the fragments of down adhering to the rough bark at the entrance. The tree is often so much decayed that a single tug at the bark near the hole will remove the whole adjacent surface, exposing the gray-green eggs where they lie in the clinging soft down. It is rather hard to locate the nest when the tree selected by the bird is in heavy timber a half mile or more from the lake, but, when the female is sitting, it may be done by making an early morning trip to the lake, remaining under cover, and waiting for her to come to the lake to feed. She generally arrives between 9 and 11 and immediately joins the drake. After splashing and preening her feathers, she feeds most industriously for perhaps an hour and then flies directly back to the nest.
I include here data for three nests taken in the Okanagan region.
Okanagan, British Columbia, May 12, 1916. A nest containing eleven fresh eggs was found in the hayloft of a deserted log barn, on the shore of a lake. The eggs were placed in a hollow scooped in the straw under a heavy beam which rested on the piled-up straw. The loft was well lighted through ?he spaces between the logs and by a large opening at one end. This situation is, of course, most unusual, but it had apparently been used some years before the nest was found. I had seen broods of young on this lake in previous years, when I was not able to find the nest. The birds would generally alight on top of a chimney in an unused house close by before flying into the barn.
Farney's Lake, Okanagan, May 31, 1912. A nest with seven partly incubated eggs was placed in a large cavity in a yellow pine stump, standing in eight inches of water on the shore of the lake. The cavity containing the eggs was eighteen inches above the water and the eggs were in plain view of a person standing several feet away.
Rolling's Lake, May 26, 1917. A nest containing seven fresh eggs was found in an old fir stub, standing in eighteen inches of water near the shore of the lake. The top of the stub had rotted out to a depth of two feet and the eggs were at the bottom of this cavity. Down could be seen protruding through a small hole in the stub, a few inches above the eggs.