ALFALFA BREEDING FOE DROUGHT RESISTANCE. 15 



Strain F. — This is South Dakota Xo. 164, which is thought to be 

 S. P. I. Xo. 991, a Turkestan stock. This strain is less coarse and is 

 better in quality of forage than most of the Turkestan varieties. 

 In amount of seed produced it stands second to strain E, as noted in 

 Table I. 



A part of the selections with which the breeding work was begun at 

 Belief ourche were made in 1907 by Mr. John Cole, now of the Office 

 of Dry-Land Agriculture Investigations of the United States Depart- 

 ment of Agriculture, but at that time connected with the South 

 Dakota Agricultural Experiment Station and in charge of the High- 

 more substation. These are selections A 1 and 2, B 1, C 1-4, D 1 

 and 2, E 1-7, and F 1-3. Selections E 9-16 and F 4-12 were made 

 at the Highmore substation in 1906 by Prof. W. A. Wheeler. Selec- 

 tions E 17-31 and many other selections not described in this bulle- 

 tin were made by the writer. 



BREEDING METHODS USED. 



In the alfalfa-breeding nursery (PI. I, fig. 1) plants are grown 

 singly in hills 21 inches apart, the rows being 42 inches apart. This 

 allows 75 plants to a row in the regular plats of the station, which 

 are 8 rods long. The seed from a single plant is generally planted in 

 one row of hills, but when sufficient seed was available, two rows of 

 hills have been planted to a single selection, and when the quantity 

 of seed available was small, less than a full row has been planted to 

 a selection. Where less than a row was planted there were 25 or 50 

 hills instead of 75, as in a full row. The hills are planted at definite 

 distances apart so that the rows of plants are in line in both directions. 

 (See PI. I, fig. 1.) Each row is given a progeny number and each 

 plant within the row an individual number corresponding to the 

 number of the hill in which the plant grows. If a plant is missing 

 in the row the order of numbering is not changed, each plant in the 

 row being permanently designated by the position it actually occupies. 

 This system makes a convenient and certain means of designating 

 each plant and obviates the use of stakes except at the head of the row. 



At the period when the first blossoms appear the plants in the 

 nursery are studied carefully and complete notes are taken as to 

 the type of plant, the amount of branching, leanness, and the color 

 of the flowers. The forage type of plant is best judged at this time, 

 for it is at this stage in the development of the plant that it should 

 be cut for forage. After these notes are taken all the inferior plants, 

 together with such as are divergent from the type of the row, are cut 

 and removed from the nursery. This is done in order that the pollen 

 from these inferior plants will not be carried to and fertilize the 

 flowers of the superior plants. It may be explained further that all 

 plants at the ends of the rows are discarded. This is done in order to 



196 



