N the spring of 1 9 I 6, a few weeks after the publication of Marlborough and Other Poems, a letter about the book and its author reached me from an unknown correspondent. I have had it a week, he wrote, and it has haunted my thoughts. I have been affected with a sense of personal loss, as if he had been not a stranger but my dearest friend. But indeed his personality the 'vivida vis animi' — shines so strongly out of every line, that I feel I have known him as one knows very few living people: and surely no one was ever better worth I venture to beg you, the writer went onto say, before it is too late, to give the world some fuller account of his brief us know him with his faults nothing extenuated, as his fellows knew him, with the 'rebel' side brought out — the boy who 'got not many good reports,' who was yet the same as he who stood 'with parted lips and outstretched Many other readers made the same request or urged the publication of a volume of Letters from Germany and from the Army, which had been printed privately and given to a few personal friends. At the time we were not persuaded. It seemed to us that enough had been done by publishing the poems and that, for the rest, so dear a memory need not be shared with the world. Meanwhile, however, the poems have become more widely known; curiosity is expressed about their author; and critics form impressions of his personality. If the poems are to have a place, however small, in literature, this interest is only natural and may even be welcomed; but it ought to be well-informed, and the poems alone do not give all that is needed for a true judgment. This is one reason why the present volume is published now. Anything in the way of a formal biography was not to be thought of. But in his familiar letters to his family and friends there is material enough, when taken along with the poems, for forming a picture of the writer. There is also in them a picture of the times, especially in Germany immediately before the war, and a criticism of life and literature, which may be found to have a value of their own.