The amount of manure which can be used advantageously has not been determined. It has been the general practice in California to give the young plants an abundance of stable manure, and the effect of this seems to be highly beneficial. There has been a suspicion that large amounts of manure, if applied to bearing plants, would decrease the production of fruit, but the evidence is not convincing. Lack of pollination is probably the cause of many crop failures which are attributed to excessive soil fertilization.

Plants of the compact low-growing type require almost no pruning. Those of tall straggling form often need cutting back in order to keep the branches from developing to such great length that they cannot support their own weight.

Seedling feijoas do not reproduce the parent variety and are less satisfactory than plants propagated by some vegetative means. Layering is used in France. In the United States many plants have been grown from cuttings, and not a few by whip-grafting.

When seedlings are grown, they should be from plants which produce good fruits in abundance. If kept dry, feijoa seeds will retain their viability a year or more. One of the best mediums for germinating them is a mixture of silver-sand and well-rotted redwood sawdust. They are small and delicate, and should not be planted in heavy soil. A light sandy loam, containing much humus, is satisfactory. The seeds should be sown in pans or flats, covering to the depth of } inch. Germination usually takes place within three weeks. A glasshouse is not necessary, but the flats containing the seeds should be kept in a frame with lath or slat covering to provide half-shade. As soon as the young plants have made their second leaves they should be pricked off into two-inch pots; after attaining a height of 4 inches they should be shifted into three-inch pots, from which they can later be transplanted into the open ground.

Layering is somewhat tedious, but with the feijoa is more successful than any other vegetative means of propagation. Those branches which are nearest the ground are bent down and covered with soil for the space of 3 to 6 inches. They require no care except keeping the soil moist. They will root in about six months, after which time they may be severed from the parent and set in their permanent positions.

Cuttings are successfully rooted under glass, and occasionally in the slat-house or lath-house. They should be of young wood from the ends of branches, and about 4 inches in length. Inserted in clear sand over bottom-heat, they will strike roots in a month or two; without bottom-heat they root very slowly. It is sometimes advised to keep them covered with a bell-jar. In Florida good results have been obtained by using as cuttings the young sprouts which appear around the base of the plant; these are removed with a heel when still quite small, and are planted in sand. Although they are slow to form roots, the percentage of loss is lower than when branch-tips are used.

Whip-grafting has given good results in some instances, and is probably one of the best methods of propagating the feijoa. The stock-plants should be of the diameter of a lead-pencil, the cions slightly smaller and of firm wood. Grafting has been successful both under glass and in the open ground.

Many feijoa plants which have been grown in California have borne little or no fruit. It has commonly been thought that wrong cultural practices were the cause of this, but the investigations of K. A. Ryerson and the author indicate that self-sterility may be to blame in many instances.

In its native home, the feijoa is believed to be pollinated by certain birds that visit the flowers in order to eat the fleshy sweet petals. The stamens and style project to a considerable height in the center of the flower; they brush against the breast of the visiting bird and pollen-grains adhere to its feathers. When it visits the next plant some of these pollen-grains are likely to come in contact with the stigmas of other flowers and remain upon them. Cross-pollination is thus effected.

In the United States the birds which do this work in the habitat of the feijoa are not present; consequently the plant must depend on other pollinating agencies. In some instances feijoa plants are self-fertile, and abundant fruits are produced when the flowers are self-pollinated. In other instances, it has been found that they are self-sterile, and can develop fruits only when pollen is brought from a different plant. The pollen of self-sterile feijoas has been found potent, when applied to flowers of other individuals.

To avoid the dissemination of self-sterile feijoas, varieties known to be self-fertile should be propagated by vegetative means. Seedlings, even if grown from a self-fertile variety, may nevertheless be self-sterile.

Grafted or layered plants begin bearing two or three years after they are planted. Seedlings may not bear until the fourth or fifth year. Self-fertile varieties often yield regularly and abundantly. The ripening season in California is October to December. The fruits fall to the ground when mature, and must be laid in a cool place until they are in condition for eating, - which can be known by their becoming slightly soft, and by their perfumed aroma. They spoil quickly in a hot, humid atmosphere, but if stored in a cool place they may be kept a month in good condition. They can be shipped long distances without difficulty. Feijoas are usually packed for market in fruit-baskets holding about two quarts.

To be appreciated, this fruit must be eaten at the proper degree of ripeness. M. Viviand-Morel says, "Everyone knows that the finest pears are only turnips if eaten a trifle too soon or a trifle too late." The observation is applicable also to the feijoa.

The plant is attacked by few insect pests. The black scale (Saissetia oleoe Bernard) is the principal enemy which has been noted. No fungous parasites have yet become troublesome.

In the Pomona College Journal of Economic Botany (February, 1912), the writer has described three varieties of the feijoa, the Andre, the Besson, and the Hehre. The Andre, described below, is the only one which has been widely disseminated. Other varieties which have originated in California as seedlings have been propagated to a limited extent, but they are little known as yet.

Andre. - Form oblong to oval; size medium, length 2 to 2\ inches, breadth 1 1/2 inches; base rounded, the stem inserted without depression ; apex rounded, the calyx-segments cupped; surface roughened, light green in color, overspread with a thick whitish bloom; flesh whitish, juicy, of spicy, aromatic flavor suggesting the pineapple and the strawberry; seeds few, small. Season November and December on the French Riviera and in southern California.

This variety is of unknown origin. It was brought to France from Uruguay in 1890 by Edouard Andre, and was planted in his garden at Golfe-Juan, on the Riviera. Layered plants were later sent from France to California. It is self-fertile, and fruits profusely. The shrub is sometimes erect and open in habit, and in other instances low, compact, and broad.