AN OLD saying that goes something like "one for the cutworm, one for the crow, one for the damp and one for to grow," may have been a useful rule of thumb before technology took much of the risk out of planting seeds.

But, with high seed costs, including fees for some of that technology, farmers may want to cut back a bit on plant populations to save money on crops that are increasingly expensive to grow.

And they can justify the reductions, according to one year's worth of Texas Experiment Station data.

Todd Baughman, Extension agronomist at Vernon, is evaluating Bollgard cotton seeding rates to determine the most profitable plant population.

Baughman explained his findings recently at the Munday, Texas Experiment Station field day.

Baughman said new transgenic cotton varieties offer a lot of potential to many Southwest farmers, but higher seed costs and technology fees increase expenses.

"We wanted to see if we could find ways to make the technology more affordable," he said. "The first thing we thought of was seeding rate. I was surprised at the results from the first year."

Baughman said a test evaluating final plant population, yield and gross return for various seeding rates indicated that the lowest rate provided the best return.

"We tested Bt cotton at two, four, six, eight, 10 and 12 seed per foot of row in 1999. We cut out the 10 and 12 seed tests for 2000 because we saw no advantage to that heavy a rate from the first year," he said.

Baughman used a John Deere vacuum planter to assure accurate seed placement. They counted plants just before harvest to determine which seeding rate offered the best survival.

"With two seed per foot of row, we averaged 1.9 live plants at harvest," Baughman said. "At four per foot, we averaged 3.9 live plants. At 12, only eight live plants had survived.

"We also counted barren plants, plants without a harvestable boll. At two and four seed per foot of row, we found less than 5 percent barren plants; at six per foot, we found 8 percent and from eight to 12 seed per foot, 12.5 percent of the plants were barren. A barren plant is a weed. It does nothing to help production."

Baughman said a significant number of seeds at higher seeding rates provided no economic advantage.

Plant height at the lower seeding rates (two to four per foot) increased some 10 inches.

"We did not treat any of the test plots with Pix," he said. "Adding Pix will allow farmers to regulate plant height."

Lower seeding rates also produced the highest yields. "We averaged 1,300 pounds per acre with two seed per foot of row," Baughman said. "Last year was a tough year for cotton, but we irrigated for maximum yields."

He said the four and six-seed plots also produced well. "Anything over six decreased yield."

Baughman said seed cost savings could be significant with lower rates. At two seed per foot, seed costs run about $19 per acre. That jumps to $55 per acre with six seed.

"Our highest gross returns came from two seed per foot of row," Baughman said. "Farmers will benefit by cutting seeding rate with transgenic varieties."

Baughman cautions that data represents only one year of research and does not recommend seeding rates as low as two. "I'm not certain that two is the magic number," he said. "I think the best rate will fall somewhere between two and four, probably two or 3.5 per foot of row."

He's currently collecting data from the 2000 test, also an extremely dry year, and will repeat the experiment one more time.

"I would like to see results in a more typical rainfall year," he said.

Baughman planted his test plots on 40-inch rows, about six or seven pounds of seed per acre, to achieve two per foot, 13 to 14 pounds for four per foot.

With narrower rows, extend the distance between seed but aim for 26,000 to 52,000 plants per acre, he said.

He said two to four seed per foot of row should work well for both irrigated and dryland cotton.

"But if growers are planting under adverse weather conditions, they may not want to drop all the way down to two seed per foot. Accurate seed placement also is more critical when lowering seeding rates that far," Baughman said.