Cones were collected in the fall of 1984 from trees in a stand of balsam fir in Canaan Valley, West Virginia, and seed was planted at the Ohio Division of Forestry nursery at Marietta, Ohio, in the spring of 1985. In the spring of 1989, trees were lifted from nursery beds and then planted at two sites - the Ohio Agricultural Research and Development Center's (OARDC) Pomerene Forest Laboratory near Coshocton, Ohio, and at OARDC's Wooster, Ohio, campus. Weed control in plantings included preplanting spraying of three-foot-wide strips with glyphosate, followed by annual applications of simazine and atrazine (2 pounds a.i.a. each).
Two series of nitrogen fertilization studies, involving five sites, were carried out. In the spring of 1993, ammonium nitrate applications at 0, 100, 200, and 300 pounds of actual N per acre were made on plots at the Wooster and Pomerene sites identified earlier, with follow-up applications at the same rates in 1995 and 1996. Physical and chemical properties of soils for the two study sites (Wooster 1 and Pomerene 1) are given in Table 1. There was considerable variation in internal soil drainage on plots, both between and within replications. At the Pomerene 1 site, soils were well- to moderately well-drained, with loam to silt loam surface textures and silt loam to silty clay loam subsoil textures. At the Wooster 1 site, they were moderately well to somewhat poorly drained, with silt loam surface textures and silt loam to silty clay loam subsurface textures. Optimum chemical soil properties for the West Virginia sources of balsam fir have not been established, but pH, P, K, Ca, and Mg levels at the Wooster site are at, or above, levels that have been identified as being desirable for other Christmas tree species grown in Ohio, including the more exacting species such as the spruces and Douglas-fir. At the Pomerene Forest Laboratory, soil pH, Ca, and Mg levels are somewhat low, but probably not critically so. Phosphorous levels are more than adequate, but K levels are below desirable levels for even the less exacting species such as Scotch pine. At each site, a randomized complete block experimental design was used, with four replications and 10 trees in each individual plot.
| Table 1. Chemical and Physical Properties of Soils on Nitrogen Fertilization Study Sites. | ||||||||
|---|---|---|---|---|---|---|---|---|
| Study | Chemical Properties | Physical Properties | ||||||
| Location | Year Estab. | pH | P | K | Ca | Mg | Depth. Mottl. ins. | Surface Soil Texture |
| ---lbs/acre--- | ||||||||
| Wooster 1 | 1993 | 5.9 | 49 | 179 | 1400 | 232 | 11-18 | silt loam |
| Wooster 2 | 1996 | 6.0 | 82 | 248 | 1440 | 208 | 9-13 | silt loam |
| Pomerene 1 | 1993 | 5.4 | 48 | 62 | 860 | 117 | 18-36+ | loam/silt loam |
| Pomerene 2 | 1996 | 5.3 | 52 | 85 | 860 | 138 | 36+ | sandy loam/loam |
| Pomerene 3 | 1996 | 6.0 | 25 | 72 | 1300 | 216 | 0-5 | silt loam |
In the spring of 1996, additional studies were established on three sites (Table 1) - a somewhat excessively well-drained area (Pomerene 2) having sandy loam to loam surface and loam subsurface textures; a poorly drained site (Pomerene 3) having silt loam surface and loam to silt loam subsurface textures; and a somewhat poorly drained site (Wooster 2), having silt loam surface and loam to silty clay loam subsurface textures. Chemical properties of soils at the Wooster site and the "dry" site at Pomerene were similar to those for the 1993 study plots; for the "wet" site at Pomerene, soil pH was considerably higher (6.0) than for the 1993 study area. In those studies ammonium nitrate was applied at rates of 0, 100, and 200 pounds of actual N per acre using a randomized complete block experimental design, with three replications and seven to 10 trees per plot.
Trees in all studies were measured yearly, before trees were sheared, for terminal shoot growth, needle lengths, and numbers of lateral limbs on the terminal shoot. Uniform four-inch-long foliage samples of current year's shoots were also collected from the upper one-third of tree crowns in mid-September of each year, oven dried at 70 degrees C, and weighed, and those weights were used as one measure of response to fertilization. Needles from those samples were separated from twigs and analyzed for total N by the Research-Extension Analytical Laboratory at OARDC's Wooster campus.
Visual, qualitative color evaluations were made in late December of the upper and lower surfaces of needles. In those evaluations (all made by the author), a twig sample was selected at each study site from a tree having an upper-surface color at the yellow-green midpoint (3.0) on a 5-point "visual" scale: 1 = yellow, 2 = green-yellow, 3 = yellow-green, 4 = green, 5 = blue-green. That sample was compared with the upper surface of needles of each tree in the plots, and a rating was assigned. A similar procedure was used for the lower surface of needles using a 3-point scale: 1 = slightly bluish, 2 = moderately bluish, 3 = bluish.
Analyses of variance (for randomized complete block experimental designs) were done for each individual study (two at the Wooster site and three at the Pomerene site) by OARDC's Computing and Statistical Services using means of measurements for the seven to 10 trees in each plot or composited foliage samples from each plot as items. When analyses of variance indicated statistical significance at the 5 percent probability level, a least significant differences (LSD) test was used for mean separation.