172 



BEACTION-INTENSITIES OP STARCHES. 



continued closeness. In Iris iberica (Charts D 646 and 

 D 666) there is a difference in the type of the two curves 

 in the pyrogallic-acid reaction, but not in the chloral- 

 hydrate reaction, and in the former the curves tend to 

 marked separations, but in the latter to marked closeness. 

 In Phaius grandifoUiis (Charts D 655 and D 667) the 

 same peculiarities are observed. Similar pairs of charts 

 of the curves of other starches with these and other 

 reagents exhibit corresponding characteristics. It is 

 of importance to recognize that the differences be- 

 tween the two curves may be as marked in the 

 reactions of the same starch with different reagents 

 as it is in the case of different starches with the 

 same reagent. Indications of these differences have 

 had incidental reference in the immediately preceding 

 statements, and they may be sufficiently accentuated by 

 reference to a single generic group of reactions, as, for 

 instance, the reactions of Iris iberica with different rea- 

 gents (Charts D 668 to D 688), that which is found here 

 being taken as a rough index or suggestion of the records 

 of the other starches. 



3. Composite Reaction-intensity Curves with 



DlFFEEENT AgENTS AND EeAGENTS. 

 (Charts E 1 to E 46, and D 1 to D 691.) 



In the construction of the composite reaction-inten- 

 sity curves the abscissaa are, in the polarization, iodine, 

 gentian-violet, and safranin reactions in terms of gross 

 quantitative light and color values based on an arbitrary 

 scale of 105 in divisions of twentieths; in the tempera- 

 tures of gelatinization, in the centigrade scale in divisions 

 of 2.5° ; and in the reactions with the chemical reagents 

 on a duplex scale, the upper portion giving the time of 

 complete or practically complete gelatinization (95 per 

 cent or more of the total starch), and the lower portion 

 of the scale the percentage of total starch gelatinized 

 when complete or practically complete gelatinization has 

 occurred within not less than an hour. The ordinates 

 represent the agents and reagents used in the reactions. 

 The reaction-intensity of each agent and reagent is 

 marked upon its ordinate and upon the proper abscissa, 

 and then a line is continued from ordinate to ordinate, 

 making an irregular curve. This form of chart is espe- 

 cially useful in the differentiation and recognition of 

 varieties, species, subgenera and genera, and in compari- 

 sons of the peculiarities of parents and hybrids. The 

 method of construction is, however, faulty, and the curves 

 are at times misleading because differences that have 

 been recorded antecedent to the record used in the chart 

 may be of very different significance, on which account 

 there will be found here and there what appear to be 

 discrepancies from what should be expected upon the 

 basis of the data of the systematist; but as previously 

 stated, each of these different kinds of charts brings 

 out in a particular way certain features, and it is of pri- 

 mary importance to note that there are presented in 

 Charts D 1 to D 691 data of the progress of the reactions 

 that are of essential importance in connection with 

 understanding and proper interpretation of these com- 

 posite charts. In a word, the composite charts exhibit 

 in a gross and by no means accurate way comparative 

 reaction-intensities. For instance, the reaction-intensi- 

 ties of two or more starches may be shown to be 95 per 

 cent of the total starch gelatinized in 30 minutes, or pre- 



cisely the same, whereas the records for the preceding 

 periods may or may not have shown any differences. 

 This is illustrated in the uranium-nitrate reactions of 

 Amaryllis belladonna, Phaius grandifolius, and Miltonia 

 vexillaria (Chart D689), wherein at the end of the 

 5-minute period the figure for both Amaryllis and Phaius 

 is the same or 65 per cent; and that of Milt'onia 83; 

 and at 15 minutes, and thence onward, they are practi- 

 cally exactly the same for all three. Then again, the 

 curves of gelatinization of any given starch may undergo 

 a complete change in its relationships to other curves 

 during its progress. This is well shown in the cobalt- 

 nitrate reactions with the same starches (Chart D 690). 

 At the end of the 5-minute period the order of reactivity 

 is Miltonia, Amaryllis, and Phaius; at 15 minutes, 

 Amaryllis, Miltonia, and Phaius; and at the end of the 

 30, 45, and 60 minute intervals, Amaryllis, Phaius, and 

 Miltonia. 



In making the composite charts the records of these 

 species at the end of 60 minutes are taken, and quite a 

 different impression is given of relative reaction-intensi- 

 ties than if the records had been used at the 5- or 15- 

 minute periods. Another source of fallacy is to be found 

 in the tendency in most of the reactions for convergence 

 or divergence of the curves, this being apparent not only 

 in the charts of the reactions of the starches of parents 

 and hybrid, but also when the curves of arbitrarily 

 selected starches are compared. This latter is set forth 

 in the pyrogallic-acid reactions of the Amaryllis, Phwms, 

 and Miltonia starches (Chart D691). Here it will be 

 noted that while the Miltonia curve is highest, that of 

 Amaryllis lowest, and that of Phaius intermediate, at 

 the end of the 5-minute period the figures are 50, 6, and 

 5 per cent, respectively; at the end of the 1 5-minute 

 period 34, 40, and 72 per cent, respectively; at the end 

 of the 30-minute period 50, 75, and 84 per cent, respec- 

 tively; and at the end of 60 minutes 94, 90, and 67 per 

 cent, respectively. In a word, at the end of the 5-minute 

 period there was no practical difference between Amaryl- 

 lis and Phaius, but a wide difference between them and 

 Miltonia; and during the progress of the reactions, while 

 gelatinization in Phaius tends to keep about parallel in 

 intensity with that in Miltonia, that in Amaryllis tends 

 to approach more and more closely the intensity of reac- 

 tion in Miltonia, so that by the end of the hour the 

 figures for Miltonia and Amaryllis are very nearly the 

 same (94 and 90 per cent, respectively) while the figure 

 for phaius is only 67 per cent. Notwithstanding the 

 grossness of this method of charting and the manifest 

 tendency to introduce fallacies, it will be apparent by 

 even a cursory survey of these charts from the aspect of 

 taxonomy that they are not without very considerable 

 value, and that by necessary modifications in the plan of 

 charting we shall arrive at a positive means by 'which 

 plants can be identified and classified by the physico- 

 chemical peculiarities of their starches and other complex 

 metabolites, in other words, by a strictly scientific 

 method. 



In Publication 173 similar charts were presented. In 

 their formulation the number of reactions was less, the 

 reagents somewhat different from those used in the pres- 

 ent research, and the values expressed were in terms of 

 complete or practically complete gelatinization time. At- 

 tempts were made in the present investigation to lessen 



