It is possible that this scientific method sometimes makes its author too bold a lawgiver. The error of most of the books is in attempting too little and in doing that little ill. They are all written for beginners only. The error of "Newport" lies in too absolute an adherence to principles. His "theory of double points" is excellent, but his theory of "the right of declining" is an innovation all the more daring because it is so methodically put. The principle has long been familiar, though never perhaps quite settled, that where two distinct points were made by any stroke,--as, for instance, a bridge and a roquet,--the one or the other could be waived. The croquet, too, could always be waived. But to assert boldly that "a player may decline any point made by himself, and play precisely as if the point had not been made," is a thought radical enough to send a shudder along Pennsylvania Avenue. Under this ruling, a single player in a game of eight might spend a half-hour in running and rerunning a single bridge, with dog-in-the-mangerish pertinacity, waiting his opportunity to claim the most mischievous run as the valid one. It would produce endless misunderstandings and errors of memory. The only vexed case which it would help to decide is that in which a ball, in running the very last bridge, strikes another ball, and is yet forbidden to croquet, because it must continue its play from the starting-point. But even this would be better settled in almost any other way; and indeed this whole rule as to a return to the "spot" seems a rather arbitrary and meaningless thing. The same adherence to theory takes the author quite beyond our depth, if not beyond his own, in another place. He says that a ball may hit another ball twice or more, during the same tour, between two steps on the round, and move it each time by concussion,--"but only one (not necessarily the first) contact is a valid roquet." (p. 34.) But how can a player obtain the right to make a second contact, under such circumstances, unless indeed the first was part of a _ricochet_, and was waived as such? And if the case intended was merely that of ricochet, it should have been more distinctly stated, for the right to waive ricochet was long since recognized by Reid (p. 40), though Routledge prohibits, and Fellow limits it. Thus even the errors of "Newport" are of grave and weighty nature, such as statesmen and mathematicians may, without loss of dignity, commit. Is it that it is possible to go too deep into all sciences, even croquet? But how delightful to have at last a treatise which errs on that side, when its predecessors, like popular commentators on the Bible, have carefully avoided all the hard points, and only cleared up the easy ones! _Poetry, Lyrical, Narrative, and Satirical, of the Civil War._ Selected and Edited by RICHARD GRANT WHITE. New York: The American News Company.