REACTION-INTENSITIES WITH EACH AGENT AND REAGENT. 



149 



difference between the two curves in each is nearly the 

 same; both are higher in the second and third than in 

 the first; and the curve in all three is lower than in 

 Amaryllis and Brunsvigia. In Hcemanthus the curves 

 are inverted, the temperature curve being the lower, 

 and the distance between the curves is practically the 

 same. In the Crinums the curves recross, the tempera- 

 ture curves being the higher, and the distances between 

 the curves in the three species are quite different — in the 

 two hardy species the distances are small but different, 

 and in the tender species well marked, showing definite 

 subgeneric division. In the three ITerines, in the first 

 the temperature curve is the higher, and in the second 

 and third the lower. In other words, Nerine crispa has 

 a higher reactivity in the temperature than in the chloral- 

 hydrate reaction, while N. bowdeni and JV. sarniensis 

 var. corusca major exhibit the opposite peculiarity. 



These remarkable inversions and reversions, both in- 

 tergeneric and intrageneric, have been found to be com- 

 mon in the researches with the various reagents, as will 

 be seen. In Narcissus the temperature curve is again 

 the higher, and in Lilium inversion again occurs, the 

 temperature curve in all four being the lower, the dis- 

 tance between the two curves being very marked in the 

 first species, marked in the other three, and nearly the 

 same in each. In 7m the temperature curve is the 

 higher in the first, third, and fourth, and lower in the 

 second; and the distance between the curves is different 

 in each, it being greatest by far in the fourth. In both 

 Gladiolus and Tritonia the temperature curve is the 

 higher, and the difference between the two curves is small 

 and practically the same in both genera. In Begonia 

 inversion again occurs, in both the temperature curve 

 being lower and very markedly lower than the chloral- 

 hydrate curve, the separation being greater in Begonia 

 socotrana. In Phaius- crossing again occurs, and again 

 in Miltonia, the separation in the former being distinct 

 and in the latter marked. While the courses of these 

 curves vary greatly, the variations are not more than 

 in the temperajture-pyrogallic acid and temperature- 

 nitric-acid charts (Charts B 5 and B 6), or when the 

 temperature curve is compared with that of any other 

 of the reagents, or when the curves of almost any two 

 reagents arbitrarily selected are compared. 



Comparisons of the temperature-pyrogallic acid and 

 temperature-chloral hydrate charts (B 5 and B 4) bring 

 out many striking differences: The range of reaction 

 intensities of pyrogallic acid is distinctly greater than 

 with chloral hydrate; the temperature and pyrogaUic- 

 aeid curves show far less tendency than the temperature 

 and chloral-hydrate curves to any relationship in their 

 courses; the variations in the degrees of separation in 

 the temperature and pyrogallic-acid curves bear no evi- 

 dent relationship to what was seen in the temperature- 

 chloral hydrate chart; and the points of inversion and 

 recrossing of the curves have no correspondence unless 

 of apparently a purely accidental character. The tem- 

 perature-chloral hydrate reactions with Amaryllis and 

 Brunsvigia show only small differences between the two 

 curves, the temperature curve being the lower in Amaryl- 

 lis and the higher in Brunsvigia; and in the temperature- 

 pyrogallic acid reactions the temperature curve is the 

 lower in both, and there is extremely little or practically 



no separation in Amaryllis but marked separation m 

 Brunsvigia. In the former, in Hippeastrum, the tem- 

 perature curve is the higher, while in the latter it is the 

 lower, and the manner of separation of the curves is very 

 different. In the former, in Hcemanthus, the tempera- 

 ture curve is the lower ; in the latter, in the first species 

 it is the higher and in the second species the lower, and 

 the differences in the degree of separation are very 

 different. In the former, in Crinum, the temperature 

 curve is the higher in all three species ; in the latter, it 

 is the lower in all three, and the separations of the 

 curves wholly unlike. In the former, in Nerine, the 

 temperature curve is the higher in one and the lower 

 in two ; in the latter, it is higher in all three ; and while 

 the chloral-hydrate curve is high in the former the pyro- 

 gallic-acid curve is very low, almost zero, in the latter. 

 In both the former and the latter charts, in Lilium the 

 temperature curve is the lower, and there are some dif- 

 ferences in the separation of the curves. In Iris and 

 throughout the remainder of the charts similar differ- 

 ences will be found. Comparing now the temperature- 

 nitric acid chart (Chart B 6) with the foregoing, it will 

 be seen that it presents a very different picture, and 

 here also there are the vagrant variations in the degrees 

 of separation of the curves and the vagrant inversions 

 and reversions, but which do not bear more than acci- 

 dental relationships to the variations observed hereto- 

 fore. In other words, each chart presents evidence in 

 suj)port of certain well-defined principles regarding 

 reactive intensities of different starches with different 

 reagents, and is a specific and characteristic picture that 

 is indicative of the particular reagent. 



From the point of view of strictly fair comparisons of 

 the temperature and chemical-reagent reactivities some 

 fallacy is introduced, because these two groups of reac- 

 tivities have not an identical basis of valuation, and 

 therefore because the value expressed by the space be- 

 tween any two abscissEe in the temperature reactions may 

 not have the equivalent values of reagent reactions. In 

 constructing the temperature scale in this research ad- 

 vantage was taken of data obtained in the previous in- 

 vestigation, and the scale was made to include what 

 was believed to be the lowest and highest temperatures 

 of gelatinization of the kinds of starches thart; were 

 Hkely to be studied, this scale being taken to be the 

 equivalent in values of the scale of reaction-intensities 

 with reagents that was made to extend between the ex- 

 tremes of highest and lowest possible reactivities. But 

 it will be seen, upon examination of Charts B 4, B 5, and 

 B 6, that the temperature reactions are limited in the 

 starches examined between 55.8° {Lilium tenuifolium) 

 and 83° {Hamanthus hatherinos) ; whereas, in the 

 chloral-hydrate reactions the values extend between 5 per 

 cent of the total starch gelatinized in 60 minutes 

 {Crinum zeylanicum) to 99 per cent in 10 minutes 

 {Begonia single crimson scarlet), and in both the pyro- 

 gallic-acid and nitric-acid reactions the values vary prac- 

 tically from extreme to extreme of the scale. 



The temperature scale as thus constructed represents 

 a scale that has Just about one-half the abscissae values 

 represented by the chemical-reagent scale. If now the 

 former scale is modified so that the extremes represent 

 the extreme temperatures recorded among the starches 



