312 DIFFERENTIATION AND SPECIFICITY OF STARCHES. 



menter, and the time-reactions of the chemical reagents may vary enormously, according 

 to the standard adopted. At the outstart of the investigation the determination of the 

 intensity of the iodine reactions was attempted by means of a calibrated color-scheme, 

 but this was found valueless because of the differences in color, as, for instance, a tint of 

 violet could not properly be matched with one of blue. If all starches reacted in tints 

 of the same color, the method would have been easy enough. Then again arose the trouble 

 of different grains of the same starch reacting wath different degrees of intensity; as a 

 consequence, an arbitrary standard was fixed which expresses the degrees very light, light, 

 fair, dark, and very dark, the average degree of coloration being taken as the index. 



In recording the times of the reactions there are opportunities for considerable errors, 

 owing chiefly to the heterogeneous mixture of grains, some grains reacting immediately, 

 others perhaps not for minutes, and others in turn not for an hour or hours. In some 

 starches, with a given reagent 90 per cent of the grains wiU be broken down in 4 or 5 min- 

 utes and the remaining 10 per cent within 60 minutes; and in another starch the reaction 

 may go on quite regularly, so that at the end of an hour it likewise is complete. In both 

 instances one hour is required for complete gelatinization, but ob\-iously the reaction was 

 much more intense on the whole in the first than in the last. In most starches there is a small 

 percentage of grains (usually from about 0.25 to 1 per cent) that are very distinctly more 

 resistant than the others ; hence it is usually necessary to disregard these grains in recording 

 the time the reaction has reached its termination. Therefore, in most of the records the 

 time of completion of the reaction means that all, or practically all, of the grains have 

 been disorganized. 



Another source of error is the presence of impurities within and without the grain. 

 The former can not be eliminated without introducing sources of fallacy greater than the 

 impurities themselves; and the latter, as in the case of most starches that must be prepared 

 from dry seeds, sometimes offer difficulties that are not easy or which are impossible to 

 meet without injury to the chemical peculiarities of the starch-grain itself. 



It must therefore be obvious, with such gross methods and such heterogeneous material 

 to investigate, and where, as in the present preliminary study, usually but a single experi- 

 ment is made with each starch with each reagent, that the results must be regarded as being 

 of a grossly quantitative or qualitative character, and absolutely tentati^'e until such inves- 

 tigations can be carried out under conditions which will yield figures that can safely be 

 adopted as fixed standards or constants. In fact, it is probable that their greatest value 

 lies in iiicUcating what unportant results are to be expected by research cai'ried out along 

 strictly scientific lines. 



With so many contributory factors to fallacious results, the reader, as was the author, 

 may be skeptical as to what value, if any, is to be attached to results obtained under such 

 conditions. He, like the author, is dependent upon the minds and hands of others (see 

 Introduction, page 14), and like the author he can make liis own deductions from the 

 records, which, if they have intrinsic value, must not only speak for tliemselves but speak 

 plainly and satisfactorily. That they have a high intrinsic value will be apparent to any 

 one who will compare the records of horticultural forms and varieties of a given species, of 

 closely related species, of distantly related species, of closely related genera, etc. Such 

 remarkable relationships, like or unhke, in correspondence generally with botanical peculiar- 

 ities, could not occur by chance, and if in some instances there appear to be records wliich are 

 departures from what was to be expected, such differences will doubtless be found to have 

 their explanation in unrecognized errors or misinterpreted results of experiment, or in errors 

 of botanical classification, which latter unfortunately are only too numerous, as is evident 

 in the continual shifting that has been and is going on in systematic botany. In fact, 

 the general accord of the results of this investigation with those of the systematic botanist 

 is not short of astonishing. 



