SUMMARYPearl millet (Pennisetum glaucum L.) is an important grain crop for millions of poor farmers and consumers in the semi-arid region of West Africa. During the past 40 years, much research on pearl millet production practices and adoption in this region has been conducted, but an attempt to summarize these results has not been previously completed and these research results are not readily available to many West African scientists. This review was completed to address this need and integrate knowledge, and at the same time identify research needs for the future and extension priorities for semi-arid West African agro-ecological zones. Research has shown that selection of improved varieties and cropping systems, appropriate cultural practices, and recommended integrated soil, nutrient, residue and pest management can greatly increase grain and stover yields of pearl millet. However, adoption by farmers has been minimal due to limited profitability, high risk and labour demand, limited input supply, market availability and appropriate public policy. This review has 196 articles included as in-text citations (Table 1) compared to 149 articles in the reference list, indicating that only one in four articles integrated two or more topics in the research. The obvious conclusion is that most of the past research has not addressed the ‘system’ but rather one or two management practices. In addition, most studies have interpreted responses in terms of yield without addressing other important considerations for farmer adoption. Recent conservation agriculture research moves closer to addressing the larger integrative types of research needed. Such research is complex and requires sustained funding for field and laboratory activities, but also for computer simulation modelling and economic assessment.

Pearl millet [Pennisetum glaucum (L.) R. Br.] is a potential crop for the Great Plains, but there are few studies on production practices. The research objective was to determine the optimal planting time for pearl millet in Nebraska relative to sorghum [Sorghum bicolor (L.) Moench]. This would optimize pearl millet yields, thus increasing the prospect for pearl millet as a crop. Studies were conducted on Sharpsburg silty clay loam (fine, smectitic, mesic, Typic Argiudoll) and Ortello sandy loam (coarse‐loamy, mixed, mesic, Udic Haplustoll) soils between 1995 and 2001. Optimal planting times were determined by relating yields with calendar date, cumulative air heat units, cumulative soil heat units, and soil temperatures. Relative yields related to air or soil heat units were effective in determining the optimal planting time. Optimum pearl millet planting times were 399 air heat units or 410 soil heat units after 1 April for the Sharpsburg soil and 406 soil heat units for the Ortello soil. The optimal sorghum planting time was 308 air heat units or 307 soil heat units after 1 April for the Sharpsburg soil and 402 air heat units after 1 April for the Ortello soil. Both crops had large planting time windows, allowing flexibility in planting time. Sorghum outyielded pearl millet for May and early‐to mid‐June planting times by 0.57 to 2.32 Mg ha−1 while pearl millet had higher yields by 0.95 to 1.20 Mg ha−1 for late June and July planting times. Sorghum produced greater yields than pearl millet for most planting times while pearl millet produced greater yields than sorghum for very late planting times.

Grain sorghum [Sorghum bicolor (L.) Moench] is the major cereal crop used in the Central Plateau of Burkina Faso to produce the traditional beer called "dolo". Experiments combining five water management techniques (WMT) and four fertilizer treatments (FT) in a randomized complete block design with a split-plot arrangement of treatments were conducted from 2003 to 2005. Water management techniques were allocated to main plots and FT to subplots. The objective was to determine the best cropping practice to optimize yields for two red grain sorghum varieties, IRAT9 and ICSV1001 (Framida), commonly used for dolo production. Results indicated that water conservation using tied-ridges produced higher grain yields. The highest yield benefit was 241 kg ha -1 for Framida. In the IRAT9 field, the highest yield benefit of 395 kg ha -1 occurred in the lowest rainfall year (736 mm) of 2005. In all years, microdose consisting of application of 19 kg N ha -1 , 19 kg P ha -1 and 19 kg K ha -1 as complex fertilizer NPK at planting, with addition of 20 kg P ha -1 as triple super phosphate at planting in the planting hole and 30 kg N ha -1 as urea applied 45 days after planting (microdose + 20 kg P ha -1 + 30 kg N ha -1 ) produced the highest grain yield increases from 420 to 756 kg ha -1 for Framida and from 812 to 1346 kg ha -1 for IRAT9. In the IRAT9 field, microdose + 20 kg P ha -1 + 30 kg N ha -1 produced the highest grain yield in all WM treatments, with yield increases from 518 to 1327 kg ha -1 depending on the WMT. Correlations indicated that the number of panicles harvested ha -1 were associated with grain yield for the two varieties. The best cropping system to optimize grain yield of Framida and IRAT9 was the use of tied-ridges and application of microdose + 20 kg P ha -1 + 30 kg N ha -1.

No abstract