Chlorophyll meters are used as a quick, inexpensive method of estimating leaf N concentration in both experiments and production fields. Direct use of the meter readings is complicated by effects of crop age and cultivar on leaf N concentration, at least partly due to variation in leaf thickness. Research in rice (Oryza sativa L.) shows that readings can be adjusted to account for these effects, and this study sought to establish whether similar relationships exist for tropical maize (Zea mays L.). Additionally, we examined several aspects of sampling methodology. In several field trials in Mexico in 1994 and at different stages of growth (6, 8, 14 leaves expanded) and for plots with different N treatments (0 N, 21 cultivars; 150 kg N h−1, 7 cultivars), meter readings were taken from five leaves per plot, with five readings near the middle of the leaf blade. Leaf N concentration (g N kg−1 dry matter) was significantly linearly correlated with the meter readings (y = 1.46x − 30.68, r2 = 0.81), but the coefficients of fit differed greatly across data sets and growth stages. Adjusting the meter readings by dividing by specific leaf weight (SLW, g leaf m−2 leaf) resulted in an improved fit within and across data sets (y = 33.47x − 6.55, r2 = 0.97). The meter readings were also directly correlated with specific leaf N (SLN, g N m−2 leaf) (y = 0.387x − 0.476, r2 = 0.92) with no adjustments. Analysis of sampling patterns determined that at least four leaves per plot are needed, with several observations per leaf, and that readings should be taken at a point lying between about 40 and 70% along the leaf length from the leaf base. For scientists and farmers with limited direct access to laboratory analysis for N, the meter provides a cheap and convenient estimate of leaf N per unit leaf area during vegetative growth.

Long-term fertility experiments with replications are often statistically analyzed as split plots in time. Years are often shown to be significantly different and the inconsistency of treatment effects over years enters into significant year-by-treatment interactions which are difficult to interpret. The objectives of this study were to evaluate long-term fertility experiments by stability analysis and relative stability and to observe possible benefits of these analyses to complement conventional analysis of variance procedures. Stability analysis which is the linear regression of treatment yield on the location/year envi· ronment mean yield was performed on long-term wheat (Triticum tu~stivum) and coro (Zea mays L.) fertility trials. Stability analysis on wheat data from the Magruder Plots, indicated that beef manure applications (269 kg N ha·') responded poorly compared to the NPK treatment when environment means were low (<2.0 Mg ha·') and visa versa when environment means were high (>2.0 Mg ha-'). Similarly, anhydrous ammonia applied as sidedressing in an irrigated coro experiment at Mead, NE, was found to be superior to ureaammonium nitrate applied either pre-plant or sidedressed when environment means were less than 8.0 Mg ha _,. Stability analysis provided a simple method of interpreting significant year-by-treatment intelilctions detected in analysis of variance models from these longterm experiments. Stability analysis may also be useful for multilocation experiments and continuous site experiments where treatments are applied to the saine plot year to year. However, stability analysis may be misleading when employed on continuous site experiments where autocorrelations are present year to year.A MAJOR PURPOSE of long-term fertility trials is to . provide a measure of the effect of the environment over time on the consistency of treatment effects. Assessing year-by-treatment interactions in long-

Surface residue accumulation under reduced and no‐till cropping systems has been shown to cause changes in the activities of several soil enzymes. The addition of crop residues containing urease increases the rate of urea hydrolysis which can increase the potential for ammonia volatilization. Our objective was to characterize the distribution of urease activity in the surface of two Oklahoma soils (Grant silt loam, fine, silty, mixed, thermic Udic Argiustoll and Norge clay loam, fine, silty, mixed, thermic Udic Paleustoll) which had been in conventional, reduced, and no‐till wheat (Triticum aestivum L.) residue management for four consecutive years. Soil cores 60‐mm deep were taken from residue management plots that had been previously fertilized with 168 kg urea‐N ha−1 yr−1 and separated into 10‐mm increments. Urease activity was determined in the soil increments and in mature undecomposed wheat straw. Urease activity was uniformly distributed in soil to a depth of 60 mm under conventional tillage. Significant increases in urease activity were measured in the surface 10 mm of both soils where accumulation of residue had occurred. Average urease activities in the surface 10 mm of soil under conventional, reduced, and no‐till residue management systems were, 12.1, 41.4, and 45.2 µg NH4‐N g−1 h−1 for Norge clay loam and 9.7, 12.4, and 35.0 µg NH4‐N g−1 h−1 for Grant silt loam, respectively. Activity of urease in mature undecomposed wheat straw grown at the site was 376 µg NH4‐N g−1 h−1 which was 28 × larger than the average activity found in soil. Wheat straw was shown to be a major source of the observed increase in enzymatic activity. Other sources for increased urease activity include increased microbial colonization during decomposition of residue and increased urea concentration in microsites due to fertilization, particularly in combination with residue.

Maize (Zea mays L.) in Central America is commonly grown under continuous cultivation and on marginal lands where restricted use of conservation practices and fertilizers limit productivity. Maize grain yield response to P source, rate, and method of application was evaluated at 33 locations in Central America. Soil orders included Andisols, Inceptisols, and Ultisols. Treatments included rock phosphate (RP) applied broadcast preplant without incorporation at rates of 13 and 26 kg P ha−1, triple superphosphate (TSP) band‐applied at planting at rates of 13 and 26 kg P ha−1, TSP broadcast preplant at 26 kg P ha−1, and a check where no P was applied. All treatments received 100 kg of N ha−1 as urea applied in split applications and 30 kg S ha−1 as CaSO4·2H2O. Averaged over locations, maize grain yield increased 0.38 and 0.74 Mg ha−1 when 13 and 26 kg P ha−1 was band‐applied as TSP and increased 0.21 and 0.16 Mg ha−1 when RP was broadcast at the same rates. All P treatments had a probability (PR) >0.61 of outperforming the unfertilized check. Triple superphosphate band‐applied at 26 kg P ha−1 offered the greatest probability (0.69) of an economic response to P fertilizer across environments. Stability analyses and analyses of yield differences suggest that response to applied P was independent of environment for the sites evaluated. The consistent response to applied P across a wide range of environments demonstrates that P is a limiting element for maize production on marginal lands in Central America and that the probability for economic response to applied P is high.

No abstract