MICHELSON] ALGONQUIAN LINGUISTIC GROUPS ABH 
There are some formations that seem thoroughly un-Algonquian; 
e. g. hédo"hok HE, SHE TOLD HIM, HER, THEM (an.), the obviative of 
which is hé#é‘hok. This formation is rare; the writer has met it but 
a few times, always in words of the same, or approximately the same, 
meaning. The stem of the examples given is hok; hé is allied with 
héi; so far as known at present there are no phonetic equivalents for 
the incorporated pronominal elements in any other Algonquian lan- 
guage. The prefixing of the termination for HE—HIM, HER, THEM (an.) 
before the initial stem is thoroughly un-Algonquian, and can not be 
paralleled elsewhere in these languages. The occurrence of the 
objective pronominal elements immediately after an initial prefix (?) 
is another anomaly. 
To sum up, Arapaho seems to have become specialized at an early 
period, but it is likely that when the phonetics of the language are 
better understood more points in common with Eastern-Central 
Algonquian will become apparent; and it is possible that borrowing 
from a non-Algonquian stock may be shown. 
EASTERN-CENTRAL 
Although the Eastern branch presents considerable differences 
from the Central branch—chiefly in the abundance of consonantic 
clusters—it is perfectly obvious that, compared with Blackfoot, 
Cheyenne, or Arapaho, it belongs intimately with the Central group. 
See the discussion of Eastern Algonquian (p. 280). 
CENTRAL SUBTYPE 
All these dialects are very intimately connected. To say that one 
dialect is not closely connected with another means merely that the 
relations between the two are not so close as between one of the 
dialects and a third. The lexical correspondence is very marked and 
the correspondence in the grammatical terminations is close. In the 
independent mode (or indicative mode) the correspondence is not so 
close as in the subjunctive. The reason for this is probably that in 
the latter case there is nothing to connect the personal endings 
with, and that in transitive forms the single pronouns (which are 
always suffixed) expressing both subject and object are so specialized 
that it is not possible readily to analyze them into their component 
elements, whereas the pronominal endings of the independent mode 
are unquestionably to be associated with the possessive pronouns 
and therefore vary more. (The Sauk, Fox, and Kickapoo forms 
in -pena, the Shawnee forms in -pe, and the Sauk, Fox, Kickapoo, 
and Shawnee forms in -pwa are wholly anomalous.) However, in 
the case of the independent mode, the analysis is far clearer than in 
