Chemical Protein Analysis: A Comparison of Kjeldahl Crude Protein and Total Ninhydrin Protein from Wild, Tropical Vegetation

Various methods are available for measuring the protein content of vegetation. This paper reviews and tests some common methods in order to provide recommendations to wildlife ecologists and primatologists. The traditional Kjeldahl crude protein method, which requires the conversion factor of 6.25 (CP6.25), was compared to values obtained through total ninhydrin protein (NP) analysis and by calculating crude protein from a newly developed conversion factor of 4.3 (CP4.3). The NP analysis gave values that were not significantly different from the CP4.3 values, and both were significantly lower than the CP6.25 values. An additional method was compared (available CP), which is CP6.25 values corrected by subtracting the acid-detergent lignin-bound crude protein. Comparisons were also made between CP4.3 and the available CP. These two methods correct the protein values differently, and theoretically CP4.3 corrects more severely, as explained in the text, but in some situations they may be correcting for the same chemical category of unavailable nitrogen. For fruits and flowers these values were not significantly different. For leaves the results were not so clear-cut. The Ugandan leaves (N = 42), Indonesian mature leaves (N = 40), Zimbabwe leaves (N = 24), and northern USA leaves (N = 11) were significantly different, and the CP4.3 always corrected more severely. The rest of the leaf sets (Belize, N = 68; Zaire, N = 36; Sumatra, N = 10; St. Catherine's Island, Georgia, USA, N = 37; and southern USA, N = 18) did not give a significant difference between CP4.3 and available CP. The choice of which analysis method to use (NP or CP4.3 versus available CP) depends on whether it seems reasonable to severely penalize all nonprotein nitrogen, considering the animal species being studied, or whether removing only the lignin-bound nitrogenous compounds is sufficient. Overall, the traditional 6.25 conversion factor is too large a conversion value for most wild vegetation; crude protein corrected only for lignin-bound protein is probably more accurate.