894 REPORT OF NATIONAL MUSEUM, 1896. 



under the caption of "Straw or ludiau Cards," from wbicli 1 have 

 extracted the following : 



To play the game a number of straws or reeds uniform In size and of equal length 

 were required. They were generally from 6 to 10 iuclies long. The number used in 

 the game was arbitrary. Lawson puts it at fifty-one, Charlevoix at two hundred and 

 one. The only essential points were that the numbers should be odd and that there 

 should be enough of them so that when the pile was divided into two parts, a glance 

 would not reveal which of the two divisions contained the odd number of straws. 

 In its simplest form the game consisted in separating the heap of straws into two 

 parts, one of whicli each player took, and he whose pile contained the odd number 

 of straws was the winner. Before the division was made the straws were subjected 



to a manipulation, somewhat after 

 the manner of shuffling cards. They 

 were then placed upon the deerskin 

 or upon whatever other article was 

 selected as a surface on which to play. 

 The player who was to make the di- 

 vision into two heaps, with many con- 

 tortions of the body and throwing 

 about of the arms, and with constant 

 utterances to propitiate his good luck, 

 would make a division of the straws 

 with a pointed bone or some similar 

 ^ instrument,' himself taking one of the 



^^.Ok '\h/ >^^X< divisions while his adversary took the 



>^ >^^ — -^ «— ^ ^^^ ^* other. They would then rapidly sep- 



arate the straws into parcels nuinber- 

 Nbl*'?^ ^°o *®o each, and determine from the 



6 fractional remainders who had the 



Fig. 206. odd number. The speed with which 



EIGHT DIAGRAMS (Tat ktcd). the proccss of couutiug was carried 



Numerical compliments indicated by numerals. On WaS alwayS a SOUrCO of WOUder tO 



the lookerson, and the fact that the 

 counting was done by tens is almost invariably mentioned. Between two people 

 betting simply on the odd number no further rules were necessary. To determine 

 which had the heap containing the odd number, there was no need to foot up the 

 total number of tens. It was to be settled by what was left over after the last pile 

 of complete tens was set aside. The number itself might be either one, three, five, 

 seven, or nine. In the more complicated forms of the games this led to giviug differ- 

 ent values to these numbers, the nine being always supreme and the one on which 

 the highest bets were wagered. It was generally understood that the holder of this 

 number swept the board, taking all bets on other numbers as well as those on nine. 

 It was easy to bet beads against beads and skins against skins in a simple game of 

 odd or even, but when the element of dilferent values for different combinations was 

 introduced some medium of exchange was needed to relieve the complications. 



'An explanation is here suggested for the origin of the familiar game of jack- 

 straws, in which a bundle of splints allowed to fall at random iu a pile are separated 

 one by one without disturbing the others. Mr. E. W. Nelson informs me that a 

 game identical with jackstraws is played by the Eskimo of Norton Sound on the 

 Yukon River, Alaska. The sticks, which are made of spruce or cottonwood, or any 

 ordinary driftwood, are about the size of a match, squared, and about four inches 

 in length. Those he collected for the U. S. National Museum were tied with a cord 

 in a bundle of about one hundred. The sticks each have the same value. They are 

 separated by means of a slender stick a little longer than the others. Another 

 method of using these sticks is to lay the bundle on the back of the baud, toss them 



