There are also considerable discrepancies in the explanation of the blanket shown in fig. 205. According to Emmons it represents a killer-whale. In each lower corner is one half of the head with teeth; right in front of the teeth, the nostril; between the two halves of the head, in the lower border of the blanket, the tail. The inverted face in the middle of the upper border, represents the body. The large square designs containing the goggle design, on each side, are interpreted as water blown out from the blow-hole. One half of the dorsal fin is indicated by a small round wing feather design in each upper corner, the human face in profile under it, representing one half of the blow-hole. According to Swanton a spirit of the sea is represented. One half of the head is shown in each lower corner, the eye design in front of the tongue being interpreted as chin. The two faces in the middle of the lower border are explained as the young ones of this sea spirit; the flicker-feather designs over them, as the inner part of the body of the old animal; the inverted large face in the middle of the upper border, as its hat; the large square designs on each side of this face, containing the goggle designs, as the dorsal fin; the two human faces in profile near the upper corners, the young ones whose bodies are shown by the round feather design over the face. Still another explanation of the same pattern was given to me. In each lower corner is shown one half of the head of a killer-whale. Its food is represented by the eye design in front of the mouth. The tail is below in the middle; the two halves of the dorsal fin are just over the tail; the inverted face in the middle of the upper border is the chest; the adjoining square designs and the attached round feather design are the flippers. Accordingly the profile faces near the upper, outer corners should be the blow-holes. [Illustration: Fig. 206. Model of totem pole with three figures representing, from below upward; sculpin, dog-fish, and sea-monster; Haida.] These examples show clearly that there is ample room for the fancy of the interpreter. The symbols lend themselves to various explanations, which are presumably selected in accordance with totemic affiliations of the owner. Not all of them seem quite consistent with the best northwest coast usage; for instance, Swanton’s explanation of the last named blanket seems doubtful on account of the inverted position of the hat and the explanation of a single wing design as a body.