30 
MUSEUM BULLETIN NO. 16. 
No further illustrations are needed. They can be found in 
any collection of North American myths. What we discover 
in every case is that the motivation of the episode is extremely 
variable and that the constant element is always the plot. 
THE MOTIFS. 
The motifs form the last unit into which the myth-complex 
can be analytically separated. Every cultural area seems to have 
a large although by no means unlimited assortment. The ex- 
treme variability with which one, then another, is used in dif- 
ferent versions of the same myth seems difficult to explain unless 
we assume that they are more or less free elements whose use 
depends in some respects on traditional association with certain 
episodes or actors, but mainly on the selective powers exercised 
by the author-raconteur and on the psychological-literary 
necessities of the plot. 
Summing up the results of our inquiry into the nature of the 
myth-complex, we may say that there are, broadly speaking, 
five units; the type of plot elaboration, the dramatis personae, 
the episodes, the motivation of the episodes, and the 
motifs; that one or more of these may vary in different 
versions of the same myth; and that, finally, the main 
problem we have to solve is to explain this variability. 
We have sought to indicate that the explanation lay in the 
manifold literary tendencies at work, particularly in the literary 
individuality of the author-raconteur. The proof of our conten- 
tion we will furnish in detail in another part of this essay. 
THE MYTH-COMPLEX AS A UNIT. 
We have discussed in the above sections the separate units 
of the myth-complex. Our separation was, however, admittedly 
arbitrary except from a purely analytical point of view. Let us 
now, therefore, look at the myth-complex as a unit. 
If we examine a number of myths, such as the Trickster 
cycle of the Winnebago, the Wrsa'ka* cycle of the Fox, the 
Nenebojo cycle of the Ojibwa, and the Coyote cycle of the Mexi- 
