450 
ANTHROPOLOGY: T. MICHELSON 
white component excepting that the closely neighboring and feeble 
jw, and are blended into a single intermediate band in each group. 
The €w series moreover is slightly shifted towards the violet. The 
positions given are the averages from five of seven negatives measured. 
In figures 3 and 4 only the groups near the reversing region have been 
indicated. Figure 5 shows all the bands thus far located in the polarized 
spectra of ammonium uranyl chloride. The diagram is by no means 
complete however for in the previous studies of the spectrum of this salt, 
already referred to, at least seven groups of fluorescence bands and eight 
groups of absorption bands were found. Preparations are in hand 
for extending the study of the absorption spectra into the ultra violet 
by means of a quartz spectrograph. 
Analogous relations between the two components of the fluorescence 
and absorption spectra of the other three uranyl double chlorides, 
2KC1. UO2CI + 2H0; 2RbCl. UO2 CI2 + 2H2O and 2CSCI.UO2CI2 +2H2O, 
have also been determined. Full data concerning these spectra and 
that of the ammonium uranyl chloride at +20° and —185° will be 
published at an early day. 
^ Polarized fluorescence, first described by Grailich {Krystall-optische Untersuchungen, 
Wien, 1858) has since been studied by Maskalyne {London, Proc. R. Soc. 28, 479), V. Lom- 
mel (Ann. Physik., Leipzig, 8, 634), E. Wiedemann (Ibid., 9, 158), Sohncke {Ibid., 58, 417); 
G. C. Schmidt {Ibid., 60, 740), and H. Becquerel, {Paris, C. R. Acad. Sci., 144, 671). 
2 Becquerel, in the paper already cited, has noted the fact that the fluorescence spectrum 
is not changed by polarizing the incident light. Sohncke obtained a similar result with 
calcite and some other crystals but found certain exceptions to the general rule. 
^ See Nichols and Merritt, Physic. Rev., Ser. 1,27,373 (1908); also Studies in Luminescence, 
Carnegie Institution (1912). 
THE LINGUISTIC CLASSIFICATION OF POTAWATOMI 
By Truman Michelson 
BUREAU OF AMERICAN ETHNOLOGY. WASHINGTON 
Presented to the Academy. May 26, 1915 
For a considerable period of time it has been commonly supposed 
that Potawatomi is very closely related to Ojibwa and Ottawa. The 
statement of William Jones in his Some principles of Algonquian word- 
formation^ is the most authoritative one on this point. In my Pre- 
liminary report on the linguistic classification of Algonquian tribes^ 
I concluded from my Hmited first-hand knowledge of Potawatomi that 
Potawatomi belonged to the Ojibwa group (comprising Ojibwa, Ottawa, 
Algonkin, Potawatomi; and somewhat removed from them Peoria, 
etc.) of Central Algonquian dialects, but that Potawatomi differed more 
from Ojibwa, Ottawa, and Algonkin than these from one another. But 
