A Biometrical Analysis of Yield Trials with Timothy Varieties Using Rod Rows 1

This paper reports yield trials with strains of timothy which were conducted at Cornell University from ~92o to !927. Twelve strains were sown in broadcast plats; and 72 strains, of which only ~2 are considered in this paper, were sown in rod rows in ~92o; and 39 strains were sown in rod rows in ~ 923Broadcast plats were ~6~ feet by 49}~ feet with pathways ~ foot wide between all plats. They were seeded with ~5o grams of seed per plat, equivalent to ~7.5 pounds per acre, on September 8, ~92o. The rod rows were ~ 7 feet long, 3 feet apart, and had pathways 3 feet wide between ends. They were seeded with o.7 ~ gram of seed per row, equivalent to 1. ~4 pounds per acre, originally estimated as ~o.3 pounds per acre regarding each row as an area of 17 feet by 4 inches. Seeding was done on September 8 and 9, z92o, and on September 15, I923. Each plat was cut as soon as convenient after its period of full bloom. Rod-row crops were bundled and dried on wire trays in the drier to air-dry weight and weighed in that condition. Broadcast plats were cut with a mowing machine and the herbage was immediately piled into crates which were weighed in the field within 5 minutes of cntting. A random sample of ~o to ~3 pounds was taken in a burlap bag, dried at ~65° to 18o°F, and used to determine the air-dry weight of the yield from the whole plat. All yields were converted to tons per acre and this unit has been used throughout the paper. For this purpose a rod row was regarded as a plat ~ 7 feet by 3 feet.