78 oe 
3. A family name constructed from a genus name logically 
implies that said genus is the most important, and that the char- 
acters of the family are, in the main, those of this genus, while in 
reality the genus may be one of the least important and materially 
differ in its structure from that of other genera, ¢. g., Cassiaceae, 
which is proposed for Leguminosaceae and Carduaceae for Com- 
positaceae. 
4. A family name should, so far as possible, describe a charac-' 
ter common to all or a great number of genera in the family, and 
this seems to have been the praiseworthy aim of several of the 
distinguished authors to whom we owe the existing nomenclature. 
The fact that plants of other families may have the same character 
does not matter; there can be no hard and fast line in classifica- 
tion. Forthis reason I prefer Cruciferaceae to Brassicaceae, Gram 
naceae to Poaceae, Contferaceae to Pinaceae, Umbelliferaceae to . 
Ammuaceae, etc., even though it is true that umbels are also found 
in Araliaceae, and cruciform flowers in Capparidaceae. 
5. Our main object being stable uniformity, the rule which 
will accomplish it with the least disturbance, and preserve the 
most of the familiar old names should be preferred ; if priority 1 
sufficient for the purpose we must reject ane more radical and sub- 
versive rule of Mr. Barnhart. 
Concerning tribal names I presume there is no question that 
they should invariably be formed from the root of a generic name 
with the addition of the termination eae, as is the usage of our 
best authors. 
Reviews. 
The 2 Life and Writings of Rafinesque siPreoared for the Filson 
Club and read at its Meeting, Monday, April 2, 1894. By Richard 
Ellsworth Call. 
Filson Club Publications, No. 10. 4to, broad margins, pp. 227- 
plates 5. John P. Morton & Co., Louisville, Ky., 1895. 
This splendid volume, creditable both from a literary and from 
an artistic standpoint, had its inception, so the preface informs 
oe, in an attempt to clear up certain matters connected with 
S ue synonymy of the Poiawides,, in which family of molluscs 
