has entirely failed, the former succeeded: two samples of Upland Georgia, 1836 and 1837; the former has partially failed; one of Bourbon succeeded perfectly. One of Egyptian 1837, entirely failed; a few seeds of Egyptian, communicated by Mr. Liddell from plants raised in his garden, some with downy, others with smooth seeds, both have vegetated: a few seeds of Pernambuco (?) gathered by myself from a tree in Bangalore, all vegetated. These examples show that much disappointment may be prevented by thus testing all foreign supplies of seed, before sowing them on a large scale.
The soil in which they are sown is a light grey sandy loam, the soil in which they are afterwards to be planted is of the same kind. But, to do justice to such an experiment, different soils should be selected, not only for the different kinds, but for each kind, as some may succeed well in some soils and fail in others. It is already known that the Bourbon and American varieties thrive in red, sandy, calcarious, also in light sandy grey, and in alluvial, soils, but altogether fail in the black cotton soils, which the country variety prefers. The Sea Island and Upland Georgia have been successfully tried on the saline sandy soils of the coast, both at Bimlipatam and Ingeram (at the latter of which places, the Bourbon had previously signally failed), and produced an abundant crop of fine cotton. The coast soils generally contain a considerable admixture of calcarious and saline matter. Such are the soils of the Sea-Islands, and so tenacious is that kind of proximity to the coast, or as it is said in America of a "saline atmosphere," that much of its good qualities are lost by removal twenty or thirty miles inland. On such, therefore, our experiments ought to be most extensively (but not exclusively) tried, and such being in some degree the character of my garden soil, it is from that and the New Orleans I anticipate the greatest success. The Bourbon however also succeeds well in grey sandy soils, as I had an opportunity of observing at Bangalore, where it seems to thrive exceedingly well.
In conclusion, I beg to add that the experiments ought not to be exclusively under the management of Europeans, neither ought the poorer ryots to be called upon to take a share in them, unless ensured against loss, but as many as possible of the wealthy and intelligent native proprietors of every district, who can be made to understand the object and principle of them, should be invited to assist in them, as their (the native proprietors') success will prove a much more satisfactory standard of comparison and means of encouragement to the poorer classes, for whose benefit principally they are undertaken, than that of Europeans, however conclusive, and will tend much more to diffuse a knowledge of the plan, and its advantages, if any, by giving them constant opportunities of
