Project Description
Sub-samples of each straw sample were prepared and retted by staff of Biolin. Each sub-sample consisted of pieces of straw 19.1 cm long cut roughly 10 cm above the ground level. Most sub-samples consisted of 100 to 150 pieces that, when combined, made a bundle about 4.1 cm in diameter. Data was collected on the unretted dry weight, the wet retted weight and the dry retted weight of each sub-sample. A reciprocating blade-type breaker/decorticator was used to roughly separate clean bast fibers from the retted dry straw and final cleaning of the bast fibers was done by hand. The weight of the clean bast fiber extracted from each sub-sample was divided by the initial dry retted weight of the sub-sample of straw to determine the percentage of bast fiber present in each sub-sample of straw. The bast fiber and shives (i.e., the non-fiber portion) produced from each sub-sample were placed in labeled plastic bags and stored for possible further analysis. The data thus generated was combined with background data supplied by PGRC on each accession and put into a computer spreadsheet format to allow easy manipulation and sorting.
Before retting, sub-samples of straw from 200 of the flax accessions were sent to the USDA Russell Research Center in Athens, Georgia for analysis using a Near Infra-Red (NIR) reflectance machine. Once wavelength patterns from these samples were generated and collected in a computer database, the straw was sent back to Biolin for retting, drying and determination of the percentage of bast fiber. After this was completed, the bast fiber and shives from each of these 200 sub-samples was once again sent back to the Russell Research Center for grinding and another round of analysis with the NIR machine. When this is completed, the database of wavelengths generated by whole and ground sub-samples will be analyzed with computer software to find which wavelength combinations are correlated most highly with the percentage of bast fiber in each sub-sample as determined by the Biolin decortication method.
The results of processing the 1,409 accessions showed bast fiber content ranging from a high of 35% to a low of 8%. From a plant breeder’s point of view, this information opens up a wide range of possibilities since it means that there is a wide range of fiber contents in the germplasm held in the PGRC collection. The results also show that newer fiber flax varieties tend to have higher fiber contents than older fiber flax varieties and that some oilseed flax varieties that were bred for Prairie conditions have fiber contents which are higher than older fiber flax varieties. The data also indicates that some commonly grown oilseed varieties have from 30-50% more fiber in their stems than other commonly grown varieties.
This data will be used by flax breeders to develop new flax varieties for the Prairies with elevated or lowered levels of fiber content and to provide base data needed to develop a NIR system that can be used to rapidly estimate fiber content in flax straw and roughly cleaned fiber. The eventual development of a NI R system to measure bast fiber content should allow buyers and sellers of flax straw and roughly cleaned fiber to improve the transparency and fairness of their commercial transactions (i.e., a farmer producing straw with a higher fiber content should get a higher price for his straw) and to provide a cheap and reliable tool that can be used to improve the fiber yield and/or consistency of the products they are producing (Le., processors should be able to quickly monitor the effect of different machinery settings on the cleanliness level of the fiber they are processing).
The reader should keep in mind that bast fiber content is only one of the characteristics that determines the amount of fiber that can be produced from a given area of land. Some of the variables which are very important but not considered here include the height of the flax plants, the number of the flax plants in a given area and the average stem diameter of the flax plants. Soil fertility, growing conditions and level of maturity at straw harvest have also been shown to have a significant influence on, not only on the fiber content, but also on fiber quality. The mandate of this project was to look only at fiber content and not at fiber quality and hence, at this time, we cannot make comments on the quality of the fiber that might be produced from different accessions.
