B tired? Give him good feed. He repays with song. Duration of sub-song. Extremely varied. Many imitations Fine perform- ance. He passes the hat. Considers what next to do. Pushes off. Roosting (Effect of wind) Thrashers are creatures of habit. Gap in B's left wing. looked ragged, tired and folorn sitting all by himself in the top of an acacia where I happened to see him. So I called him down and gave him all the worms I had with me, and then suet mixture. As he did not especially care for the mixture, I went and got soft food for him, stocking up also with worms as a precaution. He ate the soft food hungrily, then suddenly ran to low branch 10 feet from me and began quarter song. After he had been singing without pause for about 2 or 3 minutes, it occurred to me to time him. By watch the actual duration of his song was 28 minutes with only three pauses, none more than about 2 or 3 seconds long, when he had to scratch himself. Except for these, the rhythm of the song was uninterrupted. It would be inaccurate to say that he did not repeat himself during this period, but that was the impression that an uncritical observer might have gained, for he used nearly every imitation that I have heard from him(in the aggregate)and his invention seemed limitless. Except for volume of sound--a surpassing performance. At the end of the period he ran to me and ate worms. There was nothing to indicate that he was capable of making any vocal sound whatsoever. After cleaning me out, he any looked intently in all directions as if to detect evidence of affairs requiring his attention, made his decision and bolted with a strong parting push of his feet. Presumably he is directing operations somewhere on the premises. At thrasher bed-time Bb was in his usual perch and B was back in the acacia again. As there is practically a dead calm, it looks as if the wind did have something to do with his return last night to his roost behind the screen in the dorm. In the matter of night roasts it has been amply demonstrated that thrashers are creatures of habit, but that they are not slavish- ly bound by it. The gap is still present in B's left wing. If not recorded before, it should have been, that the feather is not entirely miss-