— 140 — are hard to recognise, but thoy do exist very largely in the commercial crop, and tend to produce a general deterioration in the most valuable characteristics of the crop.

It can thus be realised that an extensive depreciation in the quality and quantity of a crop may be brought about by the introduction of a single Hindi plant into a cotton area, and how that once the damage is done, years of labour must ensue before eliniination of the undesirable factors is complete.

Considerable use has been made of the target diagrams devised by Mr. W. L. Balls * on which the length of the fibre is plotted against the ginning percentage, the closeness of the pattern indicating the uniformity of the strain.

The selections made in 1913 were largely based on such diagrams, and the marked improvement in the uniformity of the strains may be clearly seen in the 1915 results.

In conclusion, it may be claimed that a definite start has been made towards the creation of a pure seed supply, for from the hundred plants chosen in 1912, sufficient seed has been obtained to sow eighteen feddans in 1915, the parental characters of each individual plant in the area being known in detail for the three intervening seasons.

Concurrently with the above, new types are constantly being isolated, so that it should be possible to obtain pure strains and meet any change that may occur in the requirements of the market.

' The Cotton Plant in Egypt."