es 26 She _ » TAssmMpBLy 
disposal of the cultivator. Hence we cannot expect to attain the same 
precision for agricultural botany as is already attained for natural bot- 
any until there is recognized that the two studies are not governed 
by the same dominant controlling motives nor by the same rules for 
classification; until it is recognized that as for the purpose of the bota- 
nist the genetical relationship of plants is the key to the proper classi- 
fication and identification of plants, so for the purposes of agri- 
cultural botany the relationship of form and use, as produced by forces 
directed by man, is to furnish the key to the agricultural classification. 
We may be pardoned for calling attention to the importance to 
agricultural experiment of a scheme which shall enable us to classify 
varieties for the purpose of identification. At present, in default of 
historical evidence of the equivalency of varieties, the work of distant, 
experimenters cannot wel] be in unison, nor can changes produced 
through cultivation or through climatic conditions or through series 
of selections be recognized and recorded in proper relation. So long 
as it is certain that the variety differences which occur in our cultivated 
plants have a distinct value to the grower, it becomes essential to accur- 
ate experimentation and to the drawing up of conclusions for practical 
directions that some method for the recognition of similar varieties 
through description alone shall be devised. ‘There is no more import- 
ant work for a station to carry on along with its more immediate 
practical duties than one which shall enable this condition of systematic 
identification of varieties to be fulfilled. 
In the study of some eleven hundred or more different varieties of 
agricultural plants as grown at'the station in 1883, some general con- 
clusions appertaining to a systematic description were obtained. 
Variations through the use of seed apparently cross-fertilized, or in 
other words, of impure variety seed, were very noticeable. 
Variation between seed planted through the use of mixed varieties 
of seed purchased. 
Variation from seeds presumably of one variety and presumably of 
not recent cross-fertilization. 
When the effect of cross-fertilization and of mixed seed was carefully 
estimated, the constancy of the plant to type, especially in that portion 
of the plant which finds use, was very marked. We are prepared to 
admit, generally speaking, that we found a great uniformity between 
individual plants of a variety in those portions for which the plant 
was grown, and a more or less variation in those parts of the plant 
considered unessential to use. ‘Thus, in beets, the tops varied more 
than did the roots; in cucumbers, the fruit varied more than did the 
foliage. 
In the older varieties of plants the constancy of the individuals to 
type was extremely well marked, even in some cases leading us to be- 
heve that the seed had resisted hybridization from adjoining varieties ; 
in the newer varieties, the so-called novelties, constancy to type was 
less well maintained. 
In general we are led to suppose that those artificial varieties, whose 
form has received recognition as desirable, display a uniformity to type 
which admits of classification and description, and a type which resists 
change so long as the plant is exposed to the conditions of cultivation. 
We are also disposed to believe that in case of such strongly marked 
