worm at a time at rather long intervals in order to keep him there Gap still in for a few minutes. Between worms he frequently preened. The gap is still in his left wing and so much of his under-down shows through 1 dove color in thin places that he looks almost dove color. He was patient for a long time, but finally wanted worms faster and took possession of the bolts, but box in my hands fearlessly. He suddenly bolted, running and flying, and began a very soft under-song barely audible at 25 feet. This contained imitations of the russet- backed thrush, tree-toad, ground squirrel, California jay and whist- ling for the dog. From this he passed directly into full song of great beauty and power without "working up to it". It contained many of the familiar phrases already recorded as well as many impossibl approximate in writing. This was followed by a repetition of soft sub-song, then full song, and so on alternately until until 5:40, at which time approaching scrapping (Nova presumably) was heard and he left to meet the new-comer. (Was this for her benefit?) Moult and song. Here is a bird in the middle of the moult, just noted as looking softest "grouchy" singing beautifully, and alternating subsong and full song with no intermediate stage. This particular series of subsongs was about the nearest approach I have heard to what might appropriately be termed the whisper song one often hears references. Human analogue. The whole performance reminded one irresistibly of a performer on a musical instrument running over his selection softly before playing Speculation it in earnest, and perhaps the same motive was present, yet the fact that Nova (presumably) was attracted by the song from some place un- known, lends plausibility to the supposition that it really intended with that end in view. On the other hand, B's attendance upon the young birds is no long- er necessary, he is not, as far as I know, building another nest, so perhaps, despite the moult and his sour looks, he may be happy after all.