During the period 1942–50 some thirty-seven multifactorial field experiments have been carried out to determine the conditions under which weeds in fibre flax can be selectively controlled by herbicides. In order to assess the direct effects of treatment on flax, some experiments have been undertaken in weed-free crops, while others have been sited in weed-infested crops so that the increases in productivity following on weed suppression could be evaluated.

In almost all experiments, the total crop weight, the yield of de-seeded straw and the amount of scutched fibre per plot were determined, while in a proportion seed yield was also measured. Where weeds were present, counts were made to establish the degree of control.

The preliminary experiments demonstrated that sulphuric acid, ammonium sulphamate and ammonium thiocyanate were far too toxic to flax. An examination of a range of cupric salts suggested that the most promising compound was cupric chloride, while of the formulations of dinitro-o-cresol investigated only the sodium salt appeared to be sufficiently selective. It was also found that in terms of fibre production sodium 2-methyl-4-chlorophenoxyacetate was less toxic than sodium 2:4-dichlorophenoxyacetate. In consequence, the bulk of the investigations have related to determining the optimal conditions for the use of cupric chloride, sodium dinitro-o-cresylate and sodium methylchlorophenoxyacetate.

For all three compounds it has been observed that flax is least liable to injury at a height range of 2–6 in. There are limitations to the amounts which can be safely employed, namely, 20 lb. of cupric chloride or 6 lb. of sodium dinitro-o-cresylate applied in 100 gal. of spray solution per acre. With sodium methyl-chlorophenoxyacetate the corresponding limit is 1 lb./acre, and the evidence suggests that the crop may be injured if the volume of application is reduced below 25 gal./acre.

Within these limitations of concentration volume and stage of development of the crop, the suppression of annual weeds has led to significant increases in the yield of straw, seed and scutched fibre. Larger gains resulted from the use of sodium methyl-chlorophenoxyacetate and sodium dinitro-ocresylate. This superiority over cupric chloride is in part attributed to the wider range of weed species which are killed by the two sodium salts and in part due to the greater liability of crop damage if the limits on amount and time of application are not strictly observed.

The results have also clearly established that for fibre flax the weight of total crop, the yield of straw, the amount of seed or plant height cannot be taken as reliable criteria for the evaluation of herbicides, for while all these may be little affected by treatment yet there may still be large reductions in the amount of recoverable fibre. A preliminary anatomical examination of the internal changes in the stems of flax has shown that when treatment with dichlorophenoxyacetic acid is delayed, a marked but localized curvature of the stem results and in this zone fibre formation is abnormal: thus breakages either during pulling or processing may occur at this point of weakness.