38 CIRCULAR 421, U. S. DEPARTMENT OF AGRICULTURE 



Native common, or South Dakota No. 12, alfalfa was more produc- 

 tive than any other variety under trial during the period 1927-30. 

 This difference was consistent from year to year. For hay pro- 

 duction there is no variety superior to the acclimated native alfalfa. 

 For seed production, varieties whose seed commands a higher price 

 are recommended. 



SHELTERBELT INVESTIGATIONS 



The need for trees to provide protection from wind and snow led 

 to the planting of the first shelterbelt at the field station. Later 

 Ardmore was selected by the United States Northern Great Plains 

 Field Station, Mandan, N. Dak., as one of four stations where a 

 definite set of uniform experimental plantings was made. These 

 experimental plantings were made to study the species of trees 

 adapted to shelterbelt use; the spacing of trees in the shelterbelt; 

 the advantages and disadvantages of pruning ; and the relative merits 

 of mulching, clean cultivation, and lack of cultivation during the 

 early years of a shelterbelt. 



During the life of the experiments nearly all the trees used were 

 shipped to the station from Mandan, N. Dak., and annual inspections 

 of the shelterbelt were made by officials from that station. 



The heavy soil and dry subsoil at Ardmore produce conditions of 

 exceptional severity, under which only the hardiest species of trees 

 were able to survive. The drought of 1931 was instrumental in com- 

 pleting the destruction of less hardy trees. 



The first shelterbelt was planted along the west side of the build- 

 ings and grounds in 1917. This belt was divided into two blocks 

 each 450 feet long, one block being spaced 4 by 4 feet and the other 

 4 by 8 feet. Siberian pea-tree in both blocks were spaced 2 feet 

 apart in the row. Nine rows of trees were planted in each block in 

 the following order from west to east : 



(1) Laurel willow (Salix pentandra L.) ; (2) boxelder (Acer 

 negundo L.) ; (3) green ash (Fraxinus pennsylvanica lanceolata 

 (Borkh.) Sarg.) ; (4) boxelder; (5) and (6) Northwest poplar 

 (Populus sp.) ; (7) American elm (Ulmus americana L.) ; (8) box- 

 elder; and (9) Siberian pea-tree (Caragana arborescens Lam.). 



All trees, except Siberian pea-tree, in the block spaced 4 by 8 feet 

 were pruned. Ordinarily the low-growing, hedgelike type, in this 

 case Siberian pea-tree, should be planted on the windward side, in 

 this locality the west side, of the shelterbelt to trap snow in the belt 

 itself, but m this planting it was placed on the east side. A good 

 stand was obtained from all species planted. 



The trees in the section spaced 4 by 4 feet did well until their 

 spread prevented cultivation with available equipment. They 

 reached that stage without shading the ground sufficiently to check 

 weed growth, and the block became a breeding plot for weeds. Weed 

 seeds produced in it were scattered to other parts of the farm. The 

 poplars made good growth, and a loss of not over 20 percent was in- 

 curred prior to the winter of 1921. They had used up nearly all 

 available moisture by autumn that year and went into the winter 

 dry. The next spring about 80 percent of them were dead, and the 

 remainder were weakened. By 1925 the only trees of this species 

 remaining were those in favored locations at the end of the row. 



