30 MASS. EXPERIMENT STATION BULLETIN 449 



of 12 mature tomatoes each were assayed. The water, total solids, total sugars, 

 and ascorbic acid contents of the tomatoes were similar for the check and the 

 mulched plots. The tomatoes from the mulched plots contained more soluble 

 solids and carotene than those from the check plots. Larger amounts of calcium, 

 magnesium, phosphorus, and potassium were found in mulch-plot tomatoes than 

 in the control-plot tomatoes. Judged by the results noted above, mulching to- 

 matoes increases the mineral, especially phosphorus and potassium, content of 

 the tomatoes. 



Studies on the Quantitative Estimation of Hemicelluloses. (Emmett Bennett.) 

 The data reported in the Annual Report for 1946-47 under this heading are pub- 

 lished in the Journal of Analytical Chemistry. 



Quantitative chemical procedures based on the removal of hemicellulose from 

 holocellulose have received further consideration. Approaches have included 

 (1) a study of the effects of acid hydrolysis, using different concentrations of 

 acids, (2) alkaline extractions at different pH levels, and (3) means to determine 

 when the holocellulose residue is substantially uniform, although possibly still 

 retaining a quantity of furfural-yielding components. Results from the first 

 and third approaches were of most promise. Limited data indicate that a slight 

 increase in the concentration of the acid is more effective in bringing about 

 hydrolysis than an increase in length of time. When plotted, these data also 

 indicate that extraction is continuous. These observations would indicate that 

 the whole cellulosic structure is attacked to some extent during the entire period 

 of hydrolysis. 



The alkali lability test, when applied to a holocellulose from which the hemi- 

 celluloses have been removed, yields an alkali number ranging from to 3; 

 before the hemicelluloses are removed by alkali, the number is of the order of 14. 

 It would therefore seem possible to use the alkali number as a guide in determin- 

 ing approximately the time at which the incrusting hemicelluloses have been 

 removed. The cellulosic residue is apparently not attacked appreciably by 

 reagents used in this test. 



The Chemical Investigations of Hemicelluloses. (Emmett Bennett.) Investi- 

 gations dealing with the chemistry of hemicelluloses have been continued, with 

 special attention to the hemicelluloses of corn stalks. Four different fractions 

 when hydrolyzed yielded approximately 55, 68, 75, and 77 percent of reducing 

 sugar as xylose. Specific rotations were all negative and had the values 40°, 

 46°, 34°, and 30°. Xylose appeared to be present in all fractions. 



Quantitative estimations of xylose as the dibenzylidene dimethyl acetal did 

 not prove successful and further work has been discontinued for the time. A 

 good quantitative procedure for xylose would be very desirable. Because of 

 certain disadvantages in the use of as-diphenylhydrazine for the determination 

 of arabinose, attempts were made to utilize benzyl phenylhydrazine. The pro- 

 cedure, while fairh' accurate for certain mixtures, was found to be unsuitable 

 for general use. 



Two of the fractions noted above yielded positive tests for starch, while two 

 were negative. In each case the starch-like substance could be removed by 

 treatment with a polidase-S enzyme preparation. Whether the presence of starch 

 has a bearing on the origin of the pentoses is still an open question. 



Results obtained thus far seem to indicate that fractions may differ quali- 

 tatively as well as quantitatively. In general, however, the fraction most resist- 

 ant to extraction is likely to be the purest and the most homogeneous in character. 



A report on the hemicelluloses of maize cobs and rye straw may be found in 

 the Journal of Agricultural Research 75: 43-47 (1947). 



