4 CIRCULAR 50, U. S. DEPARTMENT OF AGRICULTURE 



water extract and calculated either as invert sugar or as dextrose. In 

 a few cases, however, only the pressed juice was analyzed, and the data 

 in such cases refer to sugar in this juice calculated as percentage of the 

 original substance. Either copper reduction or polarimetric methods 

 were used, the former predominating. It was impossible to convert 

 all of these figures to a strictly uniform basis, particularly where the 

 reports did not give definite descriptions; but only minor differences 

 are due to this irregularity. 



ACID 



Acid is the total free acid as determined by titration against stand- 

 ard alkali. It is expressed as malic (M) or anhydrous citric (C) 

 according to which was considered to predominate, and is calculated 

 into percentage of the fresh substance. 



FUEL VALUE 



Fuel value is expressed in calories which were calculated on the 

 basis of the physiological fuel values, or 4 calories per gram of proteiu 

 and of total carbohydrates and 9 per gram of fat. 



ACCURACY AND LIMITATIONS OF THE FIGURES 



The figures here presented are the result of a careful and critical 

 study of the source material. They are as reliable and accurate as 

 they could be made with the data available. There are, however, 

 certain necessary limitations imposed on any such attempt to give a 

 composite picture of substances that vary as much among themselves 

 as do the individual fruits in any one group. The figures should be 

 used, therefore, with an understanding of their limitations. 



On account of this variability in the make-up of fruits the averages 

 of a class can be only rough estimates at best of the composition 

 of a particular sample. Some of the averages represent samples from 

 numerous types of the fruit and from various localities and growing 

 conditions, and these probably are fairly reliable as average figures for 

 their class. But in the case of some others it was impossible to get 

 data that could be regarded as representative of the class as a whole. 



Many of the data used in this study came from analyses that were 

 incomplete in the sense that not all of the constituents that are in the 

 table were determined. This will be evident from the differences in 

 the number of analyses as shown in the last line under each fruit. 

 Attention should be called in this connection to the fact that the 

 determinations of a particular constituent, as sugars for example, may 

 have come from an entirely different lot of material than that on 

 which water content was determined. This accounts for some of the 

 discrepancies. 



In general, it would be expected if all of the analyses were complete 

 that the sum of sugar, fiber, and acid would be less than the total 

 carbohydrates and, since ripe fruits have little or no starch, would 

 come within 1 to 3 per cent of it. But in some cases the total of these 

 three substances is greater, and in others it does not come within 3 

 per cent of it. Such differences may be due either to the errors of 

 sampling or to inaccuracy in analytical technic. Possibly, however, 

 there may be in some fruits undetermined substances in excess of 3 

 per cent. 



