I am personally grateful to Archibald A. Hill's experimental analysis of “The Windhover” in the December 1955 PMLA (LXX, 968–978) for the incidental enrichment of the poem offered by a fact I had not previously recognized–the fact that “dawn-drawn” may be read ‘sketched by the dawn’ as well as ‘attracted by the dawn.’ I must therefore confess myself disappointed by his flat denial of the enrichment he has offered. Mr. Hill says: “I see no way to tell whether the best substitute for ‘drawn’ is ‘attracted’ or ‘sketched.‘ The individual reader can make his own arbitrary choice, but I believe that he must choose, and can not keep both meanings. There are instances in this poem where multiple meanings must be recognized, but these occur only when there is positive evidence for more than one interpretation. Multiple meaning is not acceptable when there is merely a choice between two meanings which are both formally and structurally satisfactory” (p. 971).