29 



The most recent analysis of meals are those lately made for Kew 

 by Professor Church, of a sample received through Messrs. Lee, Crerar, 

 and Go. from Jamaica, and called "banana" meal, although it ig quite 

 possible it may have been prepared from unripe plantains. The other 

 samples were from Surinam. The latter were particularly stated to be 

 derived from the fruit of the banana (Masa sapientum.) 



Professor Church's analyses of these samples, with explanatory 

 notes are given below : — 



Professor A. H. Church, F.R.S., to Royal Gardens, Kew. 



Shelsley, Kew Gardens, 



February 16, 1893. 



Dear Mr. Morris, 



I have arranged in the table which follows, the results of my 

 analvses of some of the banana meals which you handed to me for 

 examination. The Jamaica sample is designated by the letter A. ; the 

 Surinam sample of the meal made from the interior of the fruit of Musa 

 sapientum by the letter B. ; and the meal from the peels of the same 

 truit by 0. 



Per-centage Composition of banana meals. 





A 



B 



C 





Jamaica. 



Surinam. 



Surinam. 



Water 



15.5 



14.3 



13.1 



Albuminoids (true) 



25 



2.3 



3.3 



Starch, sugar, gum, etc. 



77.7 



79.5 



58.7 



Oil 



1.0 



.7 



5.5 



Fibre 



.7 



.9 



8.7 



Ash 



2.6 



2.3 



10.7 



Nutrient-ratio 



1.32 



1.35 



1.22 



Nutrient value 



82 



83 



74 



It is noticeable how widely the nutrient ratio (or proportion of 

 albuminoids to starch plus the starch-equivalent of the oil) diverges in 

 all these meals from the ratio of the perfect food, which should show 

 the proportion of about one to five. In the analyses by Mr. L. E. 

 Asser the above divergHuee seems less marked, for he has calculated 

 the whole of the nitrogen present as if it existed in the albuminoid 

 form. I find that this is far from being the case. In this Surinam 

 sample 2 (B. above) 71 per cent, only of the nitrogen present is albu- 

 minoid, in his sample 5, made from the peels, 77 per cent. In other 

 respects my results and his agree well. 



I would further remark that sample 4 (from Jamaica) was prob- 

 ably made from fruits still more unripe than those which were used in 

 the preparation of B. and C. For in this meal no more than 56 per 

 cent, of the total nitrogen exists in the albuminoid form, the remaining 

 41 per cent, being present in the less highly elaborated state of amides, 

 etc., the food value of which is either nil or very slight. 



In all the above samples starch is more abundant than sugar ; the 

 proportion of the latter increases as the fruit ripens. 



1 



