ANALYSIS OF THE DAKOTA GHOUP FLORA. 241 



coiniiiou I'orin of L. primif/euia is recorded at Patoot by lleer and in the 

 ti( n-A of 1 1 le La-ramie Group by Ward. The type is represented in tlie Miocene 

 of Oregon and California by L. californica, passing to tlie pre.sent Laurus or 

 Persea caroliniensis by the leaves described as L. pseiido-caroliniana from 

 the auriferous gravel deposits of California. 



The species of Persea, especially P. Sehimpcfi and P. Hai/roHi, find their 

 analogy in P. pahroiiioiyha, of the flora of rxeliiiden; Liiiinis (Persea) 

 Delessli Sap., of Sezanne; L. (/randis, of Corral Hollow, California, and Per- 

 sea caroliniensis. 



Three species of Cinnamomum of the Dakota Glroup are identified in 

 more recent formations: C. Heerii\vi\\\ C. affine of the Laramie Grciup, and 

 C. poJjimorplmm of the Miocene; C. Scheuchzcri and C. sezannense, by truly 

 identical species of the Eocene and of the Tertiary. If no representative 

 of Cinnamomum is left in the present flora of North America, the absence 

 is accounted for by the same cause which has forced the disappearance of 

 species of Ficus and of a few other groups of plants which now inhabit 

 countries under the same degree of latitude but with a climate subject to 

 more moderate variations. CiHimmoDMm affine has an allied living represent- 

 ative in C. eainphora of Japan. 



Sassafras has eleven species in the flora of the Dakota (xroiip and two 

 in that of Atane. One of these, <S'. arctica, is related to S. cretaceum; the 

 other, referred to S. recurvata Lesq., of the Dakota Group flora, is founded 

 upon two fragmentary leaves scarcely identifial)le with the species. In more 

 recent formations iS'. J'faffi,a,na Heer, recorded from Patoot, is also founded 

 u\)on a fragmentary leaf. Its relation to any of the si)ecies described is not 

 well defined. Still from the Upper Cretaceous one species is recorded and 

 figured by Dawson from Vancouver Island. 



From the Laramie Group and from the Tertiary of North America no 

 otliei' species of Sassafras are recorded excej^t S. Selivi/ni Daws., from a 

 Tertiary locality near Souris River in Canada. In Europe the first leaves 

 of Sa,ssafras were described by Saporta from Sezanne as S. primigenia. The 

 affinity of this species is with Lindera Masoni of the Dakota Group. Besides 

 the species of Suzanne, only two others are recorded from the Miocene of 

 Europe, both more or less closely allied to 8. cretaceum. This last has the 

 greatest affinity with the living 6'. officinale, being represented by trilo1)ate 

 and by nearly entire leaves in the same manner as are those of <S'. offieiuale, 

 being either simple or palmately lobed. The leaves of Lindera Masoni are 

 remarkably similar to those of Lindera (Sassafras) triloba of Japan. 



MON XVII 16 



