Nidrition Investigations. 359 



For the four women who were subjects of feeding experiments, the 

 average daily amount of carbohydrate excreted in the faeces, on a 

 cellulose-free diet, estimated as dextrose, by averaging the fore-pe- 

 riods of all trials, was 0.8 gram. The utilization of carbohydrates was 

 therefore unusually good, since Atwater and Bryant'si coefficient of 

 digestibility for such a diet is 98 per cent, and undoubtedly every one 

 of these individuals consumed over 50 grams of carbohydrate per day. 

 With the exception of the subject of a single experiment who had 

 chronic constipation, these were all normal, healthy individuals, free 

 from disturbances of the alimentary tract. 



The three seaweeds fed in toto, Limu Eleele, Limu Pahapaha, and 

 Limu Lipoa, show an average digestibility of 51 per cent. This is 

 higher than that obtained in Professor Mendel's laboratory for un- 

 cooked Cetraria islandica- (average of three experiments, 15 per cent) 

 and much lower than that reported by Oshima for dried marine algae^ 

 (average 77 per cent) . 



In man, with the exception of dulse and salep, which almost entirely 

 disappeared in the alimentary tract, the average digestibility of all 

 preparations is only 34 per cent, a figure in contrast to those of Loh- 

 risch (194), who finds cellulose and hemicellulose 50 per cent digestible. 

 In dogs, the average of all preparations is 42 per cent. 



Considering that the pentosan of dulse was in a form most favorable 

 for digestion, the results with this hemicellulose are in harmony 

 with those of Konig and Reinhardt (120) who reported 75 per cent of 

 the pentosans as disappearing from the alimentary tract in man; 

 and with the averages obtained by the various investigators on ani- 

 mals, which show these carbohydrates 40-70 per cent diges- 

 tible in herbivora.* It would be desirable to repeat the experiments 

 with larger quantities, although the process of preparing the material 

 is rather laborious. It must be borne in mind, that the dulse pento- 

 san is not attacked by ordinary diastatic enzymes, but can be decom- 

 posed by soil and faecal bacteria; and although this decomposition 

 did not occur readily in pure solutions of the carbohydrate, or even 

 in a putrefying mixture, it still remains to be demonstrated whether 

 the complete disappearance from the alimentary tract is not largely 

 due to the more favorable conditions for bacterial activity w'ithin the 



^Report Storr's Agricultural Experiment Station, 1899, p. 86. 



2 Cf. pp. 297-298. 



3 Cf. p. 299. 



^ Cf. pp. 274-275. 



