by the indirect method have been useful and feeding regimens based on these data have been successful (Takeuchi et al., 1979; Cho and Kaushik, 1985; Wilson and Poe, 1985; Mangalik, 1986; Satoh et al., 1992).

This method involves measuring all the feed consumed by the fish and all of the resulting excreta. Smith (1971, 1976) and Smith et al. (1980) used an aquatic modification of the metabolism chamber designed for terrestrial animal studies; it allowed for the separate quantitative collection of gill, urine, and fecal excretions of rainbow trout. The fish were force-fed a measured amount of feed, and the various excrements were subsequently collected and analyzed for their nutrient content. The amounts of the nutrients in the excrements were then subtracted directly from those in the feed to determine the amounts retained. This method allows for determining carbon and nitrogen balances as well as DE and ME values. Also, the problem of fecal leaching is eliminated because all of the water in the chamber is included in the analyses. The method is open to criticism, however, because the fish are immobilized and force-fed and so stressed that the utilization of the feed may be compromised. Use of this method has been restricted to rainbow trout and attempts to adapt it to other species have so far been unsuccessful.

Neither the direct nor indirect method accounts for the inclusion of materials of endogenous or metabolic origin in the excreta. Therefore, the data obtained are actually apparent rather than true digestibility values. Nose (1967) found that 3.1 percent of the fecal nitrogen collected from rainbow trout came from endogenous sources, and Foltz (1978) found that the endogenous protein content of the feces rose from 3.1 to 8.4 percent as water temperature increased from 7° to 19°C. This distinction between apparent and true digestibility probably has little practical impact on feeding practices, however, and thus, the reported apparent digestibility values are quite adequate.

Very few feedstuffs are fed as the sole component of a fish diet; therefore, some researchers evaluate the digestibility of a feedstuff in combination with other ingredients in the assay diet. Cho et al. (1982) determined digestible energy and digestible protein coefficients for feed ingredients by comparing the digestibility of a reference diet with that of an assay diet that contained a mixture of the reference diet (70 percent) and the test ingredient (30 percent). The reference diet was composed of natural ingredients similar to a commercial diet. Digestion coefficients were determined for the reference and assay diets by the indirect method described above, and these coefficients were used to calculate the digestibility of the test ingredient according to the following expression:

Wilson and Poe (1985) also used this procedure to determine digestion coefficients for energy and protein in diets and diet ingredients for channel catfish. This method has advantages over testing ingredients singly in that any synergistic effect of feeding the ingredient in combination with other diet components may be realized. Also, the test ingredient may be more acceptable to the fish when fed in combination with other ingredients, which leads to a normal level of intake so that the difference between apparent and true digestibility is minimized and negative nitrogen balance is avoided.

No differences were found in protein and energy digestibility coefficients for rainbow trout as water temperature ranged from 7° to 18°C (Windell et al., 1978a; Cho and Slinger, 1979; Cho and Kaushik, 1990). The apparent lack of response to temperature change has also been found in common carp (National Research Council, 1977). Cho and Kaushik (1990) reported a small increase in crude fat digestibility at 18° versus 9°C when highly saturated animal fats were fed. The rate of movement of ingesta through the digestive tract of channel catfish was found to increase at temperatures above 26°C, but no increase in digestibility coefficients was noted (Shrable et al., 1969). It may be concluded, therefore, that nutrient digestibility is little affected by water temperature within the species' range of normal growth.

It has been found in a number of species that as meal size increases, digestive and absorptive efficiencies decrease (Kinne, 1960; Pandian, 1967; Solomon and Brafield, 1972; Elliot, 1976; Windell et al., 1978a; Andres, 1979). Wilson and Poe (1985) found that extrusion processing increased the digestibility of energy, but had no effect on the digestibility of protein as compared with pellet processing for catfish diets. Cruz (1975) found similar results with diet ingredients for catfish.

Table 4-1 presents digestibility coefficients for protein, lipids, and carbohydrates in various feedstuffs for chinook

Front Matter (R1-R12)

Overview (1-2)

1. Dietary Requirements (3-32)

2. Other Dietary Components (33-37)

3. Antinutrients and Adventitious Toxins (38-42)

4. Digestibility and Absorption (43-48)

5. Diet Formulation and Processing (49-54)

6. Feeding Practices (55-61)

7. Nutrient Requirements Table (62-63)

8. Composition of Feed Ingredients (64-71)

References (72-94)

Appendix (95-102)

Authors (103-104)

Index (105-116)