428 GENETICS IN RELATION TO AGRICULTURE 
(a) Confusion in Nomenclature.—Horticultural and agricultural plant 
nomenclature is in a state of great confusion, a condition which makes it 
necessary at the outset of scientific variety testing to make taxonomic 
studies of the known types and to adopt a system of nomenclature which 
shall be followed consistently throughout the investigations. 
(b) Inaccurate Descriptions —The type descriptions of many cultivated 
varieties are either too meager or faulty to be serviceablein critical work or 
they are wanting altogether. This often makes it necessary to secure 
seed bearing a given varietal name from various growers to make com- 
parative trials and then to choose the particular lot which may be con- 
sidered most nearly typical of the variety. Much of this expenditure of 
time and money could be saved by the general adoption of definite, 
mechanical methods of describing varieties. The Bureau of Plant 
Industry of the U.S. Department of Agriculture has advocated for years 
the wider use of variety description blanks such as the one for beans 
reproduced in Figs. 173 and 174. 
(c) Impurity of Commercial Seed.—This is one of the most serious 
difficulties in the way of accurate work in variety testing. Corbett 
states that practically every sample of seed of any of the turnip-rooted 
beets will be found to contain 50 per cent. or more of roots resembling 
more or less closely the form and other characteristics of the variety, 
but the remaining population will be made up of an admixture of all 
possible variations of the turnip-rooted beets. The condition in potatoes 
is also serious according to Corbett and before Stuart undertook by 
progeny-row and hill-selection methods to establish pure strains of com- 
mercial varieties any pretense at variety testing consisted merely in 
comparing one mixed lot with another mixed lot. 
(d) Complexity of Variation—The fact that fluctuating variations 
may be both heritable and non-heritable makes it necessary to use pedi- 
gree culture methods in any effort to compare the genetic constitution 
of varieties. 
In determining the best varieties for a given location the chief de- 
siderata are yield, quality or chemical composition and uniformity. The 
chief requisites for success are (1) an adequate system of records; (2) 
proper interpretation of the results. To facilitate the keeping of accurate 
notes about each variety the use of loose-leaf printed forms has been found 
most practicable. Three of the forms used at the Maine. Experiment 
Station are reproduced here. Fig. 175 shows a facsimile of the plot 
record blank which may be used for any crop and provides for size of 
plot, fertilization, seed used and general notes. Fig. 176 shows a fac- 
simile of the blank used for recording data on oat varieties. Fig. 177 
shows the plot index by which it is possible to trace back the pedigree of 
any plot culture as plot numbers are never duplicated. This form is 
Digitized by Microsoft® 
