Pap. No'; 24T' SHEEP ISLAND — OSBORNE, BRYAN, CRABTREE 293 



elation as to burial areas, stratigraphy, or section of the deposit ex- 

 isted. They are therefore listed below as of the whole site. Dr. 

 Arthur D. Welander identified them for us. The pharyngeal teeth 

 of cyprinid fish related to carp or squawfish, probably chub {Mylo- 

 cheilus caurinus) , are represented ; vertebrae of Cyprinidae and Catos- 

 tomidae (suckers), three caudal vertebrae of a large salmonid, 

 probably Oncorhynchus tshawytscha (chinook salmon), and miscella- 

 neous bones were represented. These bones may have been left by the 

 Indians or they may well be jetsam. 



PREVIOUS WORK, DISCUSSION, AND CRITIQUE 



Garth's paper (1952), although vitiated by a lack of the ground 

 plans, profiles, photographs, and other data, that are so necessary to 

 an understanding of an archeological manifestation, has brought forth 

 some interesting factual data. A large number of errors, many of 

 them admittedly minor, and the use of long discredited speculative 

 works together with weak or peculiar methodologies have combined 

 to present in his paper an interpretation of the middle Columbia that 

 is at strict variance with that of other workers. Inasmuch as no ad- 

 dendum to his report has appeared to rectify or point out any of these 

 conditions, it appears that this is the logical place to present (or 

 rather to re-present) the more widely held interpretations. First, 

 however, the errors in his report must be corrected or indicated : 



1. Figure 40, p. 46, tribal and archeological map. The map appears to have 

 been taken from Spier (1936). The work of Ray (1936) and Ray et al. (1938), 

 the latter published at the invitation of Spier, offer more recent and con- 

 sequently more thorough data on tribal boundaries. Garth lists Ray et aL 

 (1938) in his bibliography. Had he made use of the work outlined therein he 

 would, for example, have noted that the Lohim are a dubious quantity (ibid., 

 p. 392). 



2. Garth uses the River Basin Surveys trinomial system of site designation 

 which he assumes, I believe correctly, to be well enough known not to need 

 explanation. However, 45-BN-19, In the inset (fig, 41), is used to designate the 

 site dug by Garth, properly 45-BN (Benton) -55, Site 45-BN-19 is a small 

 superficial manifestation at the extreme lower end of the island. The Rabbit 

 Island site, 45-BN-15, is incorrectly listed by Garth as 45-WW-15 in his text 

 (p. 43) and in figure 40. Figure 41, last line, lists site 45-BN-54. Presumably 

 "53" is meant, as 45-BN-54 was never excavated. There is not the slightest 

 propriety in lumping BN-3 and -53 in analysis. BN-3 was a late contact 

 midden and burial site; BN-53 an earlier house pit site only. Data on these 

 sites could have been obtained; they have since been published (Osborne, 1957). 



3. This may be carping, but for the record it should be pointed out that 

 "islands composed of fine sand" (Garth, 1952, p, 40) are not so. All have heads 

 and cores of cobbles and gravels probably clustering about basalt outcrops. 

 Those that are known to the authors appear to have well-defined and anciently 

 developed stratigraphy of sand, gravels, clay, mixed soils, and such materials. 



4. In addition, a dubiousness concerning the acceptance of Garth's "cremation 

 floors" (pp, 41, 48) must be expressed. These were, without much doubt. 



