166 



REACTION-INTENSITIES OP STARCHES. 



ferent genera, the reaction being in some instances 

 higher, or lower, or the same, or about the same, as the 

 average reaction-intensity. In comparing the data of 

 different genera, species, or hybrids, it was usually found 

 that the two tend to fall and rise together in other 

 words, that if in one set the average mean temperature 

 of gelatinization and the average reaction-intensity is at 

 a given standard and if in the next set the temperature 

 is higher, the average reaction-intensity will be higher, 

 although the quantitative relationship between the two 

 may vary; but one may rise and the other fall, and so 

 on. The varying relationships of these two sets of reac- 

 tions will be seen by comparing the records in Table B 2 

 and Chart B 42. Strictly equivalent values in the two 

 cases are not given because the scales are different and 

 arbitrary. The range of temperature reactions are in- 

 cluded between 51.5 (Lilium parry i) and 83.25 

 (Hcemanthus konig albert), representing a range of only 

 about three-fifths of the scale, while in the reaction-inten- 

 sities, as a whole, the entire scale is included; hence, it 

 follows that strictly comparative values of the excursions 

 of the temperature curve should be amplified two-fifths. 

 This fault, however, does not interfere with the gross 

 comparisons sought. Taking the two averages for the 

 Amaryllis-brunsrigia-brunsdonna group as a starting- 

 point, it will be observed that there is a well-marked sepa- 

 ration of the two curves and that the temperature curve 

 is the lower. Both curves fall in Hippeastrum, the tem- 

 perature curve less than the other, and there is an inver- 

 sion of the positions of the two curves, the temperature 

 curve now being the higher. In Hcemanthus both curves 

 are still lower, both being close in the first set but well 

 separated and again reversed in the second set, the tem- 

 perature curve now being the lower as in Amaryllis- 

 trunsvigia-britnsdonna. This last crossing is due to pe- 

 culiarities, several times referred to, of Hcemanthus 

 puniceus. In Crinum both curves rise and undergo a 

 marked separation in the last set, the temperature curve 

 remaining in all three sets lower and changed to a less 

 degree than the other curve. In Nerine both curves fall 

 and approximate. In Narcissus the reaction-intensity 

 curve remains at the same level as in the last set of 

 Nerine, but the temperature curve rises to a point slightly 

 above the reaction-intensity curve. In all of the follow- 

 ing generic groups the temperature curve falls below 

 the other curve, the degree being very variable, and the 

 range of variability far in excess of what can be accounted 

 for by error of calibration above referred to. 



These average differences do not begin to bring out 

 or even indicate the extent and kind of these variations 

 that are found when the data for members of different 

 sets are compared. For instance, in Amaryllis-bruns- 

 vigia-brunsdonna the temperatures nf gelatinization are 

 nearly the same, the maximum difference being only 

 1.75, but the reaction-intensities vary between 76 and 

 52, the temperatures for Amaryllis and Brunsdonna san- 

 derce being practically absolutely the same, while the 

 reaction-intensity averages are 76 and 55, respectively 

 a wide difference. In other words, there may be no dif- 

 ference in the temperature of gelatinization, but a wide 

 difference in reaction-intensities. In the Crinum longi- 

 folium-moorei-poweUii set, C. pou'ellii has the lowest tem- 

 perature of gelatinization, but the highest average 

 reaction-intensity. In Iris, in the first three sets the 



temperatures are uniformly higher than in the fourth 

 set, but the relative reaction-intensities are the opposite, 

 they being very much lower in the first three sets than 

 in the last set, and the difference is proportionately far 

 more marked than in the temperatures of gelatinization. 

 In Begonia, in B. socotrana the temperature of gela- 

 tiuization is very much higher than in the other members 

 of the genus represented, but the reaction-intensity is 

 very decidedly lower. On the other hand, in Ilippeas- 

 trum the temperatures of gelatinization and average 

 reaction-intensities are in both cases very closely alike. 

 In Hcemanthus katherina the temperature of gelatiniza- 

 tion is distinctly higher than in H. magnificus, but with 

 the average reaction-intensity, although there is a tend- 

 ency, on the whole, for a starch that has a high tem- 

 perature of gelatinization to have a corresponding 

 reaction-intensity. 



In comparing the data of this table it is worthy 

 of note that while there may be evidence in some reaction 

 of a grouping of genera and of subgeueric divisions 

 there may not be in others. For instance, 'the tempera- 

 ture of gelatinization of the members of two genera may 

 be close, as in the case of Hippeastrum and Nerine, but 

 the sum and average reaction-intensities may be dis- 

 tinctly different; or the temperatures may more or less 

 distinctly individualize the genus, as in the case of 

 Lilium; or they may individualize subgeneric groups, 

 as in Iris, in which the first three sets and the last set 

 stand distinctly apart from each other. While it may 

 not be possible positively to recognize a genus upon the 

 basis of temperature of gelatinization and average reac- 

 tion-intensitiy, it is at least possible to state that it may 

 be this or that genus or positively that it can not be a 

 certain genus. For instance, having the data for Hip- 

 peastrum and Nerine, it could perhaps not be stated 

 conclusively which is which, although there is evident 

 differentiation ; but neither could possibly be confounded 

 with Amari/llis-brunsvigia, Lilium, Iris, Musa, Phaius, 

 Miltonia, or Cymbidium; nor could Lilium be mistaken 

 for 7ns or for any other genus with the exception, 

 possibly, of Cymbidium. Lilium and Cymbidium are 

 very widely separated genera, one belonging to Liliaceffl 

 and the other to Orchidacea?, and there should be a wide 

 difference in the sum-total of their reactivities, but the 

 reason why they are not here so differentiated is owing 

 to their great sensitivity to the chemical reagents. So 

 far as the temperature of gelatinization is concerned, it is 

 well established that starches obtained from very remote 

 plant sources may have the same temperature of gela- 

 tinization, which peculiarity applies also to every rea- 

 gent, both of which being in accord with what is to be 

 expected of stereoisomers. On the other hand, they may 

 exhibit differences, which vary in degree with different 

 reagents. Hence, it follows that the starches are to be 

 distinguished from each other by the collective pecu- 

 liarities of each starch compared with those of other 

 starches. 



2. VELOCITY-REACTIONS WITH DIFFERENT 

 REAGENTS. 



(Charts D 1 to D 691.) 



In the preceding section it was shown, among various 

 conspicuous phenomena, that different starches exhibit a 

 wide range of reaction-intensities with a given agent or 



