INHERITANCE THROUGH VEGETATIVE PROPAGATION. 5 



been discovered by which the peel can be removed from the fruit 

 without loss of part of the oil which it contains. Where oil was 

 determined, half of the sample was taken for the oil determination, 

 division as to color and size being made as evenly as possible. The 

 fruit for the other determinations was then pared, as little as possible 

 of the white rag being left adhering to the pulp. Both the rind and 

 pulp were weighed, and the percentage of each was calculated, any 

 loss falling where it had occurred. The pulp was passed through a 

 food grinder several times and thoroughly mixed, and samples were 

 taken for the determination of insoluble solids. Each sample was 

 washed with cold water, placed in a Gooch crucible, and dried at the 

 temperature of boiling water. 



In determining the juice content, a purely arbitrary method was 

 adopted. The pulp was considered to t>e the sum of the juice plus 

 the insoluble solids. The percentage of insoluble solids was obtained 

 by actual determination, and the percentage of juice was calculated. 

 This method gives a result somewhat higher than could be obtained 

 by using mechanical methods, but these methods are arbitrary and 

 have the disadvantage of being almost impossible of standardiza- 

 tion, so that duplication of one analyst's results is rarely possible, 

 while those obtained by several analysts vary greatly. The figures 

 s^iven, therefore, represent the theoretical quantity of juice present. 

 The best mechanical devices, unless extraction with water was used, 

 would recover scarcely 80 per cent of the quantity indicated in the 

 tables. 



Total sugar was obtained by inversion with hydrochloric acid, and 

 reducing sugars were determined by the methods of the Association 

 of Official Agricultural Chemists. The optional method, treating 

 the cuprous oxide with ferric sulphate and titrating the resulting 

 ferrous salt with potassium permanganate, was used. 



Acidity was determined by titrating the cold juice, diluted with 

 distilled water, with standard alkali solution, using phenolphthalein 

 as indicator. Kepeated determinations, titrating after boiling and 

 while hot, did not materially change the results. Apparently any 

 error due to the presence of carbon dioxide is less than that caused by 

 the difficulty of determining the end point. 



RESULTS. 



The results obtained are given in Tables 1 to 6, inclusive. The 

 description of the physical characteristics of these trees, and of the 

 yields obtained from them, may be found in the bulletins of Shamel 

 and his coworkers (o, 6) . 



Table 1 gives data derived from monthly analyses of the fruit of 

 trees representing two strains of the Eureka variety — -the Eureka and 

 the Shade Tree. 

 f 



