184 AMKRICAN MIDLAND NATURALIST 



that of nearly all if not all of the Onagraceae, it has beside this its 

 flowers in 2's, the upper part of the ovary or the stiped hypanthium 

 is deciduous in fruit, while the rest of the Evening Primrose family 

 have their flowers in 4's. Ciraca would more logically be a member 

 and type of a separate family called by Dulac Geminaceae.^ Cir- 

 caeaceae might perhaps have been a more appropriate name, 

 but Dulac did not believe that any genus or family should be 

 named after individuals. He therefore suppressed such names 

 as Circaea, Gcntiana, Hittchinsia, etc., though to be consistent he 

 must have overlooked Heracleum and Hypericum, or made effort 

 to explain them by different derivation. Type genus of family 

 is, Circaea Linn., Gen. PI. p. 24, (1754). 



Perhaps we may take occasion to describe here the following. 



Circaea alpina var. aleutica nov. var. 



Planta cum caulibus, ramis, et petiolis crassis et praecipue 

 nodis caulium et romorum intumescentibus, bracteis minutis 

 aliquibus in inflorescentia sicut in C. intermedia; stigmate florum 

 2-divisa, stylo tenui. 



Plant study with thick fleshy succulent stems, branches and 

 petioles, and swollen nodes, also minute bracts present in the 

 inflorescence. Stigma or style deeply 2-lobed. 



Type No. 256514 of the U. S. National Museum, M. W. 

 Gorman's No. 119 collected at Spacious Bay, Alaska, July 16, 1895. 



Fraxinus Linn. 



The genus Fraxinus of Linnaeus is left at the present time 

 after not a few attempts at segregation on the part of older phy- 

 tographers almost in the same condition by our manual makers 

 as is the genus Acer. As usually recognized in our floras and 

 manual of to-day the socalled "genus" Fraxinus contains plants 

 with petals and sepals, others with sepals only, others again that 

 are perfectly achlamydeous. Some of the petaliferous ashes 

 are quite choripetalous, others markedly sympetalous. This 

 means that were one to arrange a key of a flora that would work 

 satisfactorily one would have to make provision that the members 

 of this genus could by amateur botanists (for which the manuals 

 are principally intended) be determined with advantage if at 

 all, only after searching among the apetalous, choripetalous, and 

 sympetalous divisions of the dicotyledoneae. What is to be thought 



I Dulac, J. Flore Dept. Hautcs Pyr. p. 328, (1867). 



