[3032] Cato says, three and a-half.

[3033] Chalk and cow-dung. See c. [24] of this Book.

[3034] Perhaps “Tuliæ;” which would mean, according to Festus, the “cascades” or “waterfalls” of Tibur, now Tivoli.

[3035] Fée says, that if we take the word “grafted” here in the strictest sense, Pliny must have seen as great a marvel as any of those mentioned in the “Arabian Nights;” in fact, utter impossibilities. He thinks it possible, however, that a kind of mock grafting may have been produced in the case, still employed in some parts of Italy, and known as the “greffe-Diane.” A trunk of an orange tree is split, and slips of numerous trees are then passed into it, which in time throw out their foliage and blossoms in various parts of the tree, or at the top; the consequence of which is, that the stock appears to bear several varieties of blossoms at the same moment. It is not improbable that Pliny was thus imposed upon.

[3036] The plane and the oak are no longer employed for the purpose.

[3037] See B. xv. c. [25].

[3038] See c. [29] of this Book.

[3039] See B. xv. c. [17].

[3040] The mulberry is incapable of being grafted on the elm.

[3041] De Re Rust. 45. The method of planting here described is still the one most generally approved of for the olive.