12 
limitation is equally good, so far as one can judge by the limited 
genera given. He adds his peculiar method of inserting forms 
which has many advantages and is harmless even when not bene- 
ficial. 
His citations of original place and time of publication and 
type localities are excellent, and he follows a sensible rule in omit- 
ting a multitude of citations, giving only the most essential. 
is work is devoid of the stilted half-breed English so pre- 
good enough for him and ought to be for any scientific ‘ 
constant interlarding with uncouth Latin and Greek derivatives in 
the works o wh r had a classical education but came 
original author who in fact deserves the most credit, but no sys- 
tem of citation possible would do justice to all who have been re- 
sponsible for the correct names of all plants. 
The chief defect in the work is the total lack of antithetical 
arrangement and condensation. There is a skeleton key at the be- 
ginning which is good but outside of that there is nothing. The 
student has to wade through line after line of matter that is the 
same in related species. If the work were properly keyed and 
grouped so that not over three species were ever kept in the same 
group and all common characters were put in the keys and noth- 
ing in the description of the species except what was peculiar to it 
there would be no duplication and there would be great conden- 
sation without the loss of any essential. : 
In the recognition of varieties there is great variation. Salix 
longifolia is given with only a single variety while they are legion, 
while S. Sitchensis has a number of forms, all but one of Jepson’s 
creation. S. Sitchensis var. angustifolia is not properly placed and 
probably grades into S. pellita. Salix flavescens is S. Scouler- 
