parison of these methods, are more interesting, and of more importance, than the study of the relations between the finished patterns. At present we can venture to express only the belief that, while many of the methods must be the same the world over, some of them will exhibit, in every region sufficiently isolated, marked peculiarities which, if enough figures are available, will enable us with certainty to recognize their locality. Thus the methods of Caroline Islands figures cannot be mistaken for those of the Navaho Indians, or the New Guinea methods for those of the other two.
These string games may be roughly classified into those figures whereof the purpose is to form final patterns, supposed to represent definite objects; those which are tricks, wherein, after much complex manipulation of the strings, the entire loop is suddenly drawn from the hand by some simple movement; and those which are catches, wherein, when certain strings are pulled, the hand or some of the fingers may be unexpectedly caught in a running noose. Of course, there is no hard and fast
.jpg)
rule of classification; several very pretty patterns may be converted into catches.
String figures are made with a piece of string about six feet long, the ends of which must be tied together to form a single loop about three feet long. In some races a thong of skin is used; in the islands of the Pacific a cord made of cocoa nut fibre, or of human hair finely plaited, serves as a string. A woven cord which does not kink as easily as a twisted cord will prove most satisfactory; unfortunately it cannot be spliced, the ends therefore must be knotted in a small square knot or laid together and bound round with thread.
All string games begin with an opening, the object of which is to get the original loop so arranged on the hands that a number of secondary loops shall cross from the fingers of one hand to the fingers of the other, when the hands are held in what is called their usual position, namely, with the palms facing each other, and the fingers directed upward (Fig. 1).
The ninety-seven figures described on the following pages show many different openings, but fifty-seven of them begin with the same opening.
In arranging the figures, those with a common opening, and otherwise closely related, are gathered in one series, instead of being distributed into race groups; and in each series, as far as possible, simple figures are placed first.
Every finger loop has, of course, two strings, and as a rule both these strings pass between the hands to form the strings of finger loops on the opposite hand; but sometimes one or both strings of a finger loop, before crossing to the other hand,
