1919] Nelson,— Deam’s Trees of Indiana 189 
T, vorite group. The present revision is an admirable example of 
popular treatment which does not sacrifice essential scientific accu- 
racy. In all, 125 species are described, representing 49 genera in 28 
families. The term “tree” is interpreted as including all woody 
plants that usually attain a diameter of 10-15 em. — an arbitrary divi- 
sion that will not always prove satisfactory, since Hamamelis vir- 
giniana, Euonymus atropurpureus and Kalmia latifolia would seem 
to have at least as much right to be represented as Ilex decidua or 
Cornus alternifolia. Crataegus heads the list with 18 species (as might 
have been expected when it is explained that the genus was revised 
for this work by Eggleston!) and Quercus is a close second with 17. 
No less than 15 species which have been referred to Indiana 
are rejected as unconfirmed. Confirmation of included species is 
made by simple citation of collector and county of collection, without 
herbarium numbers, dates, or place of deposit. Each species is given 
a simple and untechnical description, followed by a discussion of its 
distribution, economic uses, and horticultural value; and each is 
illustrated by a careful drawing, showing a leafy branch with fruit or 
flowers, or both together. The solitary photograph of Fraxinus 
Michauzii makes us wish that the general habit of other species might 
have been illustrated in the same way. Two valuable tables are 
added, one showing the specific gravity of the wood of each species 
(ranging from 0.8372 for Hicoria ovata to 0.3164 for Thuja occidentalis), 
and the other giving measurements of 49 species, showing Platanus 
occidentalis ranking first in circumference (maximum 48 ft.) and 
Liriodendron Tulipifera in height (190 ft.). A map of the State is 
added to make the distribution more clear. The key to the families 
is constructed without regard to floral characters, and is based pri- 
marily on the leaves, so that all natural relationships are lost to view, 
as in Dr. Mosher’s recent study of the Grasses of Illinois. Doubtless 
the popular appeal of these manuals is intensified by these unscientific 
keys, but it is an open question whether systematic accuracy has not 
been sacrificed to ease of determination. 
The author makes a praiseworthy attempt to attach definite mean- 
ings to the terms ordinarily used to express degree of frequency, which 
have been sadly lacking in scientific precision. He suggests the fol- 
lowing scale: “Very common,” more than 25 trees to the acre; 
“common,” 5-25 trees to the acre; “frequent,” 1—5 trees to the acre; 
“infrequent,” 1 tree to 2-10 acres; "rare," 1 tree to every 11-100 
acres; “very rare," 1 tree to more than 100 acres; “local,” when the 
distribution is cireumscribed or in spots. While this is of course 
wholly arbitrary, it is at least a step in the right direction. 
The nomenclature is said by the author to conform “to that of the 
United States Forest Service", which means that the provinciality 
of the so-called * American" Code is unfortunately perpetuated. 
Accordingly, we are treated to such absurdities as Catalpa Catalpa 
and Sassafras Sassafras, while substitutions such as Hicoria for Carya 
