Selecting forage sorghum cultivars for silage

Eighty forage sorghum cultivars were compared in 1986 for agronomic and silage quality traits. Silage yield ranged from 5.3 to 10.0 tons (t) of dry matter (DM) per acre (mean, 7.4 t); grain yield, from 13 to 113 bushels (bu) per acre (mean, 66 bu). Percent lodging was extremely high, with a mean of 51% and only one cultivar free of lodging. Pre-ensiled DM content ranged from 22.2 to 35.4% (mean, 27.4%) and plant height from 85 to 180 inches (mean, 121 inches). Silage quality results showed that in vitro DM digestibility ranged from 44.6 to 62.1% (mean, 53%); crude protein from 4.5 to 8.2% (mean, 6.8%); neutral detergent fiber from 48.3 to 71.9% (mean, 58.4%); and acid detergent fiber from 27.1 to 49.8% (mean, 35.7%). From the 80 cultivars in 1986, 60 were selected for the 1987 trial. When compared to 1986, the 1987 means showed slightly lower silage (7.0 t) and grain (63 bu) yields and much shorter plants (93 inches). Lodging scores were dramatically lower in 1987 (10%), and DM content was higher (29.1%). The year to year effect influenced all of the agronomic traits measured. The 1986 growing season favored the early maturing forage sorghums, whereas 1987 favored the late maturing cultivars.


Introduction
Results of a 1986 Kansas State University survey of sorghum seed dealers indicated that there were more than 100 forage sorghum cultivars available in Kansas.
Prior to the interest in hybrid sorghum in the late 1950's, nearly all commercially grown forage sorghum could be traced to less than 20 introductions.Improved cultivars have been developed from a rather narrow germplasm base, with 50 to 60% of the sorghum currently grown in the United States having similar germplasm.Although two forage sorghum cultivars might have similar germplasm and appearance, their value as silage crops could be distinctly different.
Often cultivar recommendations and/or selections are made primarily on the basis of agronomic traits (i.e., silage yield, lodging score).Although silage yield is an important criterion, it is only a part of the silage resource, and it may not be the most important.A producer could be most interested in maximum silage yield for over-wintering cows.
However, for backgrounding cattle, a producer would be interested in the most economical feed cost per pound of gain and, thus, silage quality would also be an essential criterion.Choice of a forage sorghum cultivar should be based upon a number of traits, including yield potential of forage and grain, adaptation, handling and ensiling characteristics, and feeding value of the silage.
Within the forage sorghum complex there is tremendous biological variation.Our objective was to provide documentation for the agronomic and silage quality traits of many of the forage sorghum cultivars available in 1986 and 1987.

Experimental Procedures
Trial 1: 1986.From the available forage sorghum cultivars identified in the 1986 survey, 80 were selected for use in Trial 1.The selections represented a broad range of agronomic characteristics.
The cultivars were grown under dryland conditions on a silt loam soil near the Kansas State University campus in Manhattan.The plots were planted on June 4.
One month earlier, 100 lb/acre of anhydrous ammonia was applied.A soil test indicated that phosphorus and potassium were adequate.Furdan 15G ® insecticide was applied in the furrow at planting, and the following day Ramrod ® was used as the pre-emergence herbicide.In July, Cygon 400 ® was used to control greenbugs.
The cultivars were randomly assigned in a block design, each with three replicates.
Each plot was 30 ft long and four 30-inch rows wide.All grainproducing cultivars were harvested at the late-dough stage of kernel maturity.The sterile hybrids were harvested when the kernels in the unbagged heads reached the late-dough stage.Nonheading hybrids were harvested on the first frost date (Oct.15).Before harvest, all plots were reduced to 20 ft in length, and one of the two inside rows was harvested with a modified one-row forage harvester to determine silage yield.The heads from the other inside row were hand-cut, dried, and threshed to measure grain yield.The chopped forage was ensiled in 5-gallon plastic laboratory silos.Silos were opened after 90 days, and samples taken for chemical analyses.
Agronomic data collected for each plot included days to half bloom, plant height, lodging score, silage yield, and grain yield.Silage quality traits measured for each plot included dry matter (DM), crude protein (CP), in vitro DM digestibility (IVDMD), neutral detergent fiber (NDF), and acid detergent fiber (ADF).
Trial 2: 1987.From the 80 cultivars used in Trial 1, 60 were selected to use in a similar study in 1987.The plots were planted six rows wide (rather than four) and silage yield was measured by harvesting two inside rows (rather than one).All agronomic and silage data were collected by the methods used in Trial 1.

Results and Discussion
Presented in Table 44.1 are the agronomic and silage quality results of Trial 1 and agronomic results of Trial 2. For certain categories, five cultivars are identified as being extreme high and low observations.Trial 1: 1986.Silage DM yield ranged from 5.2 to 10.0 tons per acre, with an average of 7.4 tons.The cultivar with the highest yield was Funk's G 1990; the lowest was the variety, Ellis.Other high yielding cultivars were PAG 55F (9.9 t), Golden Harvest (GH) SI Gro 3 (9.6 t), Garrison Sile-all (9.3 t), and DeKalb FSla+ (9.2 t).Low yielding cultivars were Casterline Suline (6.0 t), Warner Sweet Bee Sterile (6.0 t), Ketgen KFS-2 (5.7 t), and Triumph SSll (5.6 t).
Days to half bloom ranged from 53 to 106, with an average of 71 days.The earliest cultivar was Buffalo Canex; the latest was NK Sucrosorgo 405.Five cultivars bloomed in 55 days or less, and five bloomed after 85 days.Lodging score ranged from 0 to 100%, with an average of 51 percent.Only one hybrid, McCurdy F65, did not have any lodged plants, whereas GA-TE Milkmaker, GA-TE Silomaker, and NK 326 were 100% lodged.Some plots (replicates) were so badly lodged that they were judged to be unharvestable (i e., rep 1 of Hoegemeyer 615F and rep 2 of Terra Ho-K).Plant height ranged from 85 to 180 inches, with an average of 121 inches.The shortest were DeKalb FSla+ and the variety, Rox Orange; the tallest was Pioneer 931.-Other short cultivars were McCurdy XF65, NK 300, and PAG 455.Other tall cultivars were NK Sucrosorgo 405, Funk's G 1990, Oro Red Top Kandy, and Sokota 320F.Preensiled DM content ranged from 22.2 and 35.4%, with an average of 27.4 percent.The wettest cultiver was Hoegemeyer 615F; the driest was Asgrow H8551.Only seven cultivars contained over 30% DM, and nine contained under 25% DM at harvest.
In vitro DM disappearance ranged from 44.6 to 62.1%, with an average of 53 percent.The hybrid with the highest IVDMD was Cargill 250S (a sterile); the lowest was NK Sucrosorgo 405, which also had the lowest grain yield.Only two other cultivars were 60% digestible or above; Early Sumac and Buffalo Canex; 20 cultivars were less than 50% digestible.Crude protein ranged from 4.5 to 8.2%, with an average of 6.8 percent.The NDF fraction ranged from 48.3 to 71.9%, with an average of 58.4 percent.The ADF fraction ranged from 27.1 to 49.4%, with an average of 35.7 percent.Buffalo Canex had the lowest NDF and second lowest ADF, whereas Pioneer 931 had the highest NDF and ADF fractions.
Results from Trial 1 indicated that the late maturing, nonheading cultivars had the highest silage yields.However, these high yielding cultivars (i.e., Pioneer 931 and Funk's G 1990) tended to have low IVDMD, low protein, and high fiber fractions.

Table 44 .
2. Minimum, Maximum, Mean, and Standard Deviation for the Data in Trials 1 and 2