Webber further says: "Trees which needed irrigation when the freeze came suffered rather severely, as did also trees that had been irrigated three to five days before the freeze and were thus gorged with water. The least injury seemed to be on trees that had been thoroughly irrigated two or three weeks before the freeze, and had water supposedly in what might be termed the optimum amount."

In regions subject to frosts, it is necessary to protect the trees during the first two or three winters with shelters of burlap, or by placing palm-leaves, pine boughs, or other material around them. Frost-fighting with orchard heaters is sometimes practiced where heavy frosts are expected.

A sharp frost at the time the tree is in flower may result in a crop failure, although the danger from this source is probably not great either in California or Florida, and has been over-estimated in the past. In Florida, the West Indian race usually blooms late enough to escape the coldest weather, while in California the Mexican race, though it blooms in winter, is sufficiently hardy to withstand ordinary frosts, and the Guatemalan race does not bloom until April or May. The latter race is, therefore, the safest in this respect.

In California, avocado culture is not dependent on rainfall, since irrigation is commonly practiced. In Florida, on the other hand, very few crops are irrigated, and up to the present it has been the general custom not to irrigate avocado trees, except during the first two or three summers. It is coming to be recognized, however, that a wet spring is followed by a good avocado crop and a dry one by a poor crop (a condition exactly reversed with the mango). As a result of this observation, irrigation is beginning to be practiced in southern Florida, especially in seasons when the rainfall is below normal.

The necessary soil-moisture can be supplied easily and satisfactorily, but the relative humidity of the atmosphere cannot be altered artificially; hence in regions where the humidity is exceedingly low the avocado suffers in the dry portion of the year. In Florida no attention need be paid to this subject, since the humidity closely approaches that of the West Indies and other regions where the avocado is at home. Humidity may prove, however, to be the limiting factor in parts of California. Tests in the Imperial and Coachella Valleys indicate that the trees are seriously injured by the dryness of the atmosphere. Experience shows that the Mexican race is less susceptible than the West Indian. None of the varieties so far tested, however, has proved to be so resistant to atmospheric dryness as the orange or grapefruit. Shading may help to limit the injury from this source. It has been found very beneficial in the coastal belt of California, where young avocado trees are often injured during the dry summer months by sunburn.

Another climatic factor which deserves consideration is the danger from high winds. The lower east coast of Florida is occasionally visited by a West Indian hurricane which defoliates trees, strips them of their crops, or even breaks them down. Certain parts of California are also subject to occasional high winds, less severe than the hurricane but nevertheless capable of doing much damage. To minimize the danger from this source, it is advisable to keep the trees as low as possible through pruning, since they are then much less liable to injury. The low tree has an additional advantage in that it permits of picking the fruit without the use of tall ladders, and keeps the branches more readily accessible for pruning, spraying, or thinning the fruit.

In regard to soil the avocado seems to be unusually adaptable, succeeding on the sandy lands of southern Florida, the volcanic loams of Guatemala and Mexico, the red clays of Cuba and Guatemala, the granite soils of California, and even on heavy adobe, provided the drainage is good. This question is less important, therefore, than many others connected with avocado culture. The chief requisite is good drainage.

Most of the avocado groves of southeastern Florida are situated upon limestone of the kind shown as Miami oolite. This formation comprises a narrow strip of land extending from above Fort Lauderdale on the north to some miles below Homestead on the south, being widest near the latter place, and nowhere more than thirty feet above sea level. In many parts of this region the rock comes to the surface; toward the northern end it is commonly overlaid with six inches to two feet of loose light-colored quartz sand, while below Miami the surface soil becomes very scanty, but heavier in nature, containing some clay in certain localities, and being strongly impregnated with iron, giving it a reddish color. The rock itself is soft and porous, and in the process of erosion has broken down unequally, leaving a jagged surface or the characteristic pothole formation. When first grubbed it crumbles and is readily worked, but on exposure to the air it gradually hardens, owing to the deposition of carbonate of lime following evaporation of the moisture held in the interstices.

The growth made by the avocado upon this rocky land is rather remarkable; it seems, in fact, that young orchards have done better around Homestead, where the rock comes to the surface, than they have in those areas north of Miami where there are six to eighteen inches of sand on top of the rock. The reason for this may lie in the moisture-retaining properties of this soft limestone; the roots, which are always close to the surface, here probably are kept more uniformly supplied with moisture during a period of dry weather than on light sandy soils which dry out rapidly.

The heavier Florida soils seem to be much more favorable to the growth of the tree than light sands. A yellowish or brownish subsoil in many parts of Florida indicates good avocado land. The avocado prefers a moist heavy loam, and the closer this can be approached the better will be the results.

The soils of California are probably more nearly ideal for avocado culture than any of those in southern Florida. Sandy loam, which is abundant in the southern part of the state, produces excellent growth and is giving good results. Adobe does not seem so desirable, yet good trees have been grown upon it at Orange.

Red clay has been satisfactory in Cuba and Central America, while heavy clay where well drained has produced good trees in Porto Rico.

Many problems connected with avocado culture remain to be solved. One of the most important is the adaptability of the tree to low wet lands in southern Florida. It has been the general opinion that avocados should not be planted on land where the water-table is less than three feet below the surface. Krome has observed groves on low rock-land which have been killed or badly injured by overflows, even where the water came scarcely as high as the crown roots and remained there only a few days. In several plantings on marl prairie, however, experience has been quite different. Trees on this type of land have been submerged twenty-four hours without damage to them. On the low islands along the western coast of Florida, salt water sometimes floods the groves, and this has proved fatal to many trees. It is probable, also, that the failure of one or two plantings on this coast can be attributed to the fact that the water which stands about two feet below the surface of the land is saline in character.

Until more experience has been gained regarding the adaptability of the avocado to low flat ground, occasionally subject to overflow, orchard plantings should be limited to lands where the water-table is three feet or more below the surface. In California, the best site for the orchard is a gently sloping hillside, or level ground adjacent to a slope. If of this character, and well drained but naturally retentive of moisture, the situation may be considered excellent. In regions subject to heavy winds, it is well to select a piece of ground which is sheltered by surrounding elevations.