Particularly notable are its low moisture content, its high caloric content, and its high content of vitamins A and C, minerals P, Mg, S, Ca, and Fe, and the essential fatty acids linoleic and linolenic. It is also notable for a higher protein content and for an exceptionally favorable balance among the essential amino acids, having a chemical score of 45; the most limiting amino acid was methionine, which is usual for protein from plant sources other than cereals grains. Peterson et al. (1982) noted that despite the smallness of their sample, 27 fruits from 4 clones, the nutrient composition varied considerably among the clones, suggesting that the high nutritional value of the paw-paw could be further improved by plant breeding.

Commercial quality pawpaw cultivars do not exist. The 20 extant cultivars listed in Table 1 are well suited to home production, but none exhibit the combined excellence in flavor, aroma, texture, low seed: fruit ratios, aesthetics, yields, ease of propagation, and shipping storage ability needed for commercial cultivation. The following economically relevant traits of pawpaw have been listed roughly in order of descending importance as problems: yield, fruit set, harvest methods, seediness of the fruit, perishability, (storage and handling, prolonged juvenility, propagation difficulties, aesthetics/appearance, susceptibility to pests, toxic compounds/allergens, nutrient levels, flavor and texture, and fruit size variability. The first six traits encompass the greatest problems associated with pawpaws and so are treated in greater detail below. Propagation difficulties, susceptibility to pests, and the presence of toxic compounds are also detailed.

Pawpaw yields are notoriously low, a trait they share in common with their tropical Annonaceae relatives, the Annonas (Thakur, 1965; Farooqi et al., 1970; Gazit et al., 1982). This problem is a potential obstacle to commercial development. It requires the selection of higher yielding types in order that yields may exceed the minimum needed for profitability as determined by the product price and the costs of production. Bartholomew (1962) reported the typical yield from one superior tree was 4 kg. In contrast, Ourecky and Slate (1975) reported that a yield of 11.5-23 kg from a mature tree was reasonably good, but obviously based their figure on Gould's (1939) report of .5 to 1 bushel yields, since the wording is the same and the quantities are equivalent. Little (1905) estimated that 1250-1500 trees may be planted to the hectare without crowding. Multiplying Little's figure by the Ourecky-Slate figure and assuming an inverse relationship between tree density and yield per tree generates a yield per hectare of 17,300 to 29,000 kg (9,700-11,700 kg/acre). Assuming a grower price between $1.10 and $3.30/kg, this yield equals $19,000 to $95,000 per hectare ($7,700-$38,700 per acre). (Since market prices for cultivated pawpaws do not exist, price for cherimoya (Annona cherimola Mill.), a close relative of pawpaw, was used here as a proxy pawpaw price. Frieda's Finest/Produce Specialties of Los Angeles reported (pers. comm.) that southern California cherimoya growers received $1.10 to

580